(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Two Chinese poets: vignettes of Han life and thought"

UNIVERSITY 
OF FLORIDA 
LIBRARIES 




COLLEGE LIBRARY 



Digitized by the Internet Archive 

in 2012 with funding from 

LYRASIS IVIembers and Sloan Foundation 



http://archive.org/details/twochinesepoetsvOOhugh 



TWO CHINESE POETS 
Vignettes of Han Life and Thought 



TWO CHINESE 
POETS 

J^i a netted or ^JyT^an c/^ire 
ana ^^ nouant 



BY E. R. HUGHES 



PRINCETON, NEW JERSEY 
PRINCETON UNIVERSITY PRESS 

7 9 60 



Copyright © i960 by Princeton University Press 

ALL RIGHTS RESERVED 

L.C.CARD: 59-11078 



Publication of this book has been aided by 
a grant from the Bollingen Foundation 



Printed in the United States of America 
by Princeton University Press, Princeton, N.J. 




PUBLISHER'S FOREWORD 

<><><XX>O<><XX>O0<X><><^^ 

Ernest R. Hughes was for many years Reader in Chinese 
Philosophy and Religion at Oxford University. He was the 
author of a number of Sinological studies, including Religion 
in China (with Katharine Hughes), Chinese Philosofhy in 
Classical Times ^ and The Art of Letters: Lu Chi^s ^^Wen Fuy^ 
A.D. ^0 2. Mr. Hughes was working on the manuscript of the 
present book at the time of his death in October 1956. It is 
regrettable that the work has not received the finishing 
touches that Professor Hughes would no doubt have wanted 
to make had he remained alive. Princeton University Press 
is profoundly indebted, however, to Dr. David Hawkes, a 
former pupil of Professor Hughes and now Professor of 
Chinese at Oxford University, who generously undertook the 
necessary revision of the manuscript before publication. The 
Press also wishes to express its thanks to the Bollingen Foun- 
dation for its generous assistance in support of publication. 



PREFATORY NOTE 

Mr. Hughes began this comparative study of the poems of 
Pan Ku and Chang Heng after the publication of his book 
on the Wen Fu of Lu Chi {The Art of Letters, Bollingen 
Series No. xxix, New York, 195 1). He finished it in June 
1956 when he was already suffering from the disease from 
which he died in October. The Princeton University Press 
had earlier agreed to look at the book, so it was sent to 
them. It was returned after Mr. Hughes's death, with the 
recommendation that considerable revision and additions in 
the way of notes and bibliography were required. His former 
pupil. Dr. David Hawkes (now Professor in Oxford Univer- 
sity) undertook this task, and the book as now published is 
the result of his work. 

I am deeply grateful to Dr. Hawkes for the trouble he 
has taken and the time given which he could ill spare from 
his own work, and also to Mr. Wu Shih Chang for his advice 
and help. Without these two friends my husband's last con- 
tribution to the study of Chinese literature could not have 
been put into a form in which it could be available for those 
interested in the literature of the Han period. He also hoped 
that this study of history as seen through the eyes of two 
poets would appeal to historians. 

Katharine Hughes 



PREFACE 

Some readers skip prefaces j others, perhaps advisedly, read 
them carefully. For a man like myself who has benefited 
greatly from explanatory prefaces, it seems that a debt of 
honor claims his laggard energies to write one now. In brief, 
then, in my last period of research in Peking (1933-34) there 
was a haunting feeling that for me chiseling away at "the 
Classics" — my main center of interest in Sinological studies — 
was a Sisyphean task. Each Classic had its own insoluble 
problems. To a historian in search of dateable material on 
ideological development, although a number of miscellaneous 
dates were to be found, they did not afford a basis on which 
to build: follow one main clue derived from one Classic, fol- 
low another from another Classic, and the two clues com- 
pared did not make historical sense. ... It became clear that 
a purely linguistic approach to the problems was the more 
profitable course. Some first-rank scholars were working in 
this field. Could I do anything there? An experiment or two 
demonstrated that for a lone worker in his Oxford hermitage, 
even to "settle the enclitic Se" was practically unachievable. 
In the forties I went on to the study of what the Ch^ng 
litterateurs so happily called ^^fHen wen^'* literally "double- 
harness style," first in Later Han prose compositions, then 
in the mid-Han ]u (prose poems). Their attraction lay in 
the fact that in the first-century-B.c. compendium, the Li Cht 
("Record of Rites"), there were lengthy passages, and even 
whole essays, in which there was continuous complementa- 
tion of sentences and clauses, together with tighter forms of 
sentence structure. The study was encouraging, but doubts 
assailed me when I discovered that in post-Han literature 
the acknowledged masters of style were not so much con- 
cerned with clear statement and accurate description as with 
making "delicious hanging clusters of words." I then went 
to the Han scholiasts and the historical recorders to see what 

[ ix ] 



PREFACE 

their respective styles revealed. This study on the whole re- 
assured me. I came to the tentative conclusion that by the 
first century a.d. able writers had a command of ordered 
communication which earlier ages of writers had not pos- 
sessed. 

In order to check on these various impressions I made 
a special study of Lu Chi's Wen Fu ("Prose-poem on the 
Art of Letters") written in a.d. 301-302. That study proved 
highly illuminating as to methods of approach to the classi- 
cal literature: first, find a document of manageable propor- 
tions, of certain authorship, certain date, and practically in- 
dubitable text, and then follow where it leads. This conclu- 
sion lies behind my book on the Wen Fu published in 1951 
(Bollingen Series, No. xxix. New York, Pantheon Press). 

The next step was then clear. Two acknowledged masters 
of double-harness description of objects {wu) were Pan Ku 
and Chang Heng, and within twenty to thirty years of each 
other they had composed their respective "F/^ on the Two 
Capitals." Here were dateable documents by well-known au- 
thors, the extant texts in most admirable condition. Further, 
the two authors had under contemplation the previous Han 
regime in contrast to their own. What more could an ideo- 
logical historian want in the way of reliable and illuminating 
data? In this fashion I came to these special studies, came 
to write this book. Three considerations dominate its method: 
(i) to get at the texture of the minds of the two authors 
and their contemporaries j (2) placing the authors in the 
witness box as to Han beliefs and institutions, to estimate 
their reliability j (3) to explicate the particular matters on 
which they gave evidence. 

With regard to the style of composition, it has to a large 
extent been dictated by my aim to reach not one but two 
kinds of readers: the Sinologists whose familiarity with names 
and places makes explanation unnecessary, and also a certain 
particular Western type of experienced historian who desires 
to include the Far East within his purview. That there are 

[ X ] 



PREFACE 

such, long association with historians has demonstrated to me. 
I gather they have a suspicion that, apart from the desira- 
bility of excursions outside their cultural habitat, there is 
something to be learnt from that most historically-minded 
of all peoples, the Chinese. Since I have profited so much 
in my studies by exploring these historians' minds, I cannot 
but attempt to serve their interests. Not only so: surely the 
next step in historiographical research is by appreciation of 
the merits and demerits of Chinese historiography. Also, 
since it was in the Han era that "China" first saw itself as 
a world civilization, that era above all is relevant to the 
world-conscious historian. 

Therefore, in these pages, the form of presentation in- 
volves an admixture of ABC information which the Sinologist 
does not require. Also, the more specialized documentation has 
been relegated to footnotes which in nine cases out of ten 
do not concern the general reader. One matter, which is 
possibly a moot point in discussion, is whether the Chinese 
historic culture is so unique that it stands in a class by itself. 
I have taken it for granted that in no sense is "China" a lusus 
naturae, but a phenomenon which is in all respects compara- 
ble to other such large-scale phenomena. My book seems to 
me to convey information which strengthens this impression. 
As a corollary stands the belief that the time-schedule of 
European and American cultural advance is not the arch- 
type on the basis of which value judgements can reliably 
be made. 

It remains to make grateful acknowledgement of outstand- 
ing help which I have received. First come the Chinese li- 
braries: the Tsing Hua University Library and that of the 
Provincial University of Yunnan (1942-44). Then come the 
Western libraries: the Library of Congress in Washington 
(1947), the Harvard-Yenching Institute Library (1948), the 
University of California Library, Berkeley (1949), the Clare- 
mont Colleges Library (1950-52), the Oxford Chinese Fac- 

[ xi ] 



PREFACE 

ulty Library, and the Cambridge University Library. I wel- 
come this opportunity of apologizing to the librarians con- 
cerned for straining their patience from time to time. 

Then there are Dr. Ch^en Shou-yi, in the earlier stages, 
and Mr. Wu Shih-chang in the later stages. Both these sa- 
vants have rescued me from various serious interpretational 
errors. I am also indebted to Mr. Wu for his critical inter- 
est in my problems. It has been the more generous of him 
because his own book on the history of Chinese prose is still 
in the making. To Madame Maspero I am indebted for her 
gift of a copy of Henri Maspero's Les Instruments Astro- 
nomiques des Chinois au Temfs des Han^ to Dr. Joseph 
Needham for his illuminating reactions to my questions on 
Chang Heng's type of mind and other such scientific problems. 

I regret that Dr. Hulsewe's Remnants of Han Law^ Vol. I 
(Leiden, 1955) reached me so late that its influence on my 
thinking has been only a tithe of what it doubtless will be. 
Professor Dubs' History of the Form^er Han Dynasty^ Vol. 
Ill reached me too late even to be skimmed. In regard to 
his Vols. I and II (Baltimore, 1938 and 1944), I am under 
a debt, as all students of classical China are, to him and his 
colleagues for the valuable fruits of their labors. Yet more 
am I under a debt to Dr. Tjan Tjoe-som for his Po Hu T^ung 
(Leiden, 1949 and 1952). 

E.R.H. 
Dorn Cottage 
Blockley 

Moreton-in-Marsh 
Gloucester y England 

June igs^ 

'^Melanges Chinois et Bouddhiqv.es^ VT, Bruges, 1939, pp. 183-370. 



[ xii ] 



CONTENTS 

Publisher's Foreword V 

Prefatory Note vii 

Preface ix 

Chronological Table 2 

CHAPTER I: INTRODUCTION 3 

CHAPTER II: BIOGRAPHICAL DETAILS l6 

CHAPTER III: PAN KU'S ''FU ON THE 

WESTERN CAPITAL" 2^ 

Preface. i. Introduction of the Dramatis Personae. 
2. The Topography and Geography of the Site. 3. 
The City and Its Suburbs. 4. The Imperial De- 
mesne. 5. The Emperor's Personal Dwelling- 
Places. 6. The Rear Palaces. 7. The Administra- 
tion Offices, the Archives, and Other Buildings. 8. 
The Chien Chang Palace and Its Appurtenances. 
9. The Imperial Hunt. 10. After the Hunt. 11. 
The Western Capital Guest Sums Up. 

CHAPTER IV: CHANG HENG'S ''FU 

ON THE WESTERN CAPITAL" 35 

I. The Introduction of Two Interlocutors. 2. The 
Topography and Geography of Ch'ang-an. 3. The 
Palace Buildings. 4. The Rear Apartments. 5. 
The Outbreak of Megalomania. 6. The Chien 
Chang Palace Area, with Its Spirit-Revealing Tower. 
7. The Chien Chang Area (continued). 8. The 
City and Its Suburbs with Their Inhabitants. 9. The 
Hunting Park. 10. The Imperial Hunt. 1 1. After 
the Hunt. 12. The Emperor's Undignified Amour. 
13. Final Conclusions. 14. The Nobleman Makes 
His Bow. 

[ xiii ] 



CONTENTS 

CHAPTER V: PAN KU'S "FC/ ON THE 

EASTERN CAPITAL" 48 

I. The Eastern Capital Host Jibes at the Western 
Capital Guest's Opinions. 2. Wang Mang's Usur- 
pation and the Salvation of the Country. 3. The 
Emperor Ming's Achievements. 4. The Imperial 
Hunting Preserve and the Seasonal Hunts. 5. All 
Nations Turn to Loyang. 6. The Joyful State of 
Sober Simplicity and Earnest Endeavour, with Schools 
Everywhere. 7. The Eastern Host's Final Argu- 
ment. 8. The Guest's Discomfiture. 9. The Five 
Odes. 

CHAPTER VI: CHANG HENG'S "FU ON 

THE EASTERN CAPITAL" 60 

I. The Teacher's Reaction. 2. The Real Lesson 
of the Dynastic Revolutions (Third Century B.C.). 
3. The Grossness of the Nobleman's Blindness. 4. 
The Rightness of Loyang as the Capital for True 
Sons of Heaven. 5. The New Palaces and Pleas- 
aunces Described and Appraised. 6. The Empire 
and Beyond Doing Homage. 7. The Sacrifice to 
Heaven. 8. Other Ritual Duties and Moral Alle- 
gorizing on Them: A. The Ming T'ang and the 
Five Shang Ti; B. The Ancestral Sacrifices; C. The 
Ploughing of the Field of Heaven; D. The Archery 
Rite and Its Preliminaries; E. The Mystic Arrow: 
The Significance of the Wine Festival and Enter- 
tainment of the Veterans. 9. The Later Han Hunt. 
10. The Driving Away of the Demons of Plague. 

II. The Tour of Inspection. 12. The People's Ar- 
gument with the Throne. 13. Sundry Reflections 
on Emperors and Their Way of Life. 14. Plain 
Facts of Social Economy. 15. The Nobleman 
Forced to Face Unpleasant Facts. 16. The Final 
Word from Teacher Anonymous. 17. The Noble- 
man Repents. 

CHAPTER VII: CRITIQUE OF PAN KU'S "F17 

ON THE WESTERN CAPITAL" 82 

[ xiv ] 



CONTENTS 

CHAPTER VIII: CRITIQUE OF CHANG 
HENG'S "FC/ ON THE WESTERN 
CAPITAL" 108 

CHAPTER IX: CRITIQUE OF PAN KU'S 

''FU ON THE EASTERN CAPITAL" 138 

CHAPTER X: CRITIQUE OF CHANG 
HENG'S ''FU ON THE EASTERN 
CAPITAL" 172 

CHAPTER XI: GENERAL APPRAISALS 221 

I. The Value of Pan Ku and Chang Heng as Wit- 
nesses. 2. The Body Politic Set-up: An Autocracy? 
A Theocracy? 3. The Fact of Empire and Its Ef- 
fect on Men's Minds. 4. Concerning Later Han 
Levels of Self-conscious Reflection. 5. Concerning 
the Humanitarian Ethic and Its Origins. 6. The 
Fihal Piety Strain. 7. The Spread of Educational 
Facilities. 8. The Sage-Kings. 9. Concerning Han 
"Confucianism" and the Scriptures. 10. Suggestions 
for Systematic Research. 



[ XV ] 



TWO CHINESE POETS 
Vignettes of Han Life and Thought 



CHRONOLOGICAL TABLE 

<>^<><3><>0<3K><3><><><><><>5K>^^ 

THE SHANG DYNASTY 

(founded by T'ang the Victorious) 

traditional dates 1 766-1 1 22 B.C. 

THE CHOU DYNASTY 

(founded by King Wu) traditional dates 1122-256 B.C. 

THE CH'IN DYNASTY 

(founded by King Cheng of Ch'in) 221-207 ^'^' 

THE FORMER HAN DYNASTY (founded by Liu Pang) 

206 B.C. - A.D. 8 

Emperor Kao 206-195 Emperor Chao 86-74 

Emperor Hui 194-188 Emperor Hsiian 73~49 

Empress Lii 187-180 Emperor Yiian 48"33 

Emperor Wen 179-157 Emperor Ch^eng 32-7 

Emperor Ching 1 56-141 Emperor Ai 6-1 

Emperor Wu 140-87 Emperor P'ing i B.C. - 

A.D. 5 
The Infant Emperor a.d. 6-8 

THE "NEW" DYNASTY 

(Wang Mang the Usurper) a.d. 9-22 

THE LATER HAN DYNASTY 

(founded by Liu Hsiu) 23-219 

Emperor KuangWu 25-57 Emperor Shun 126-144 

Emperor Ming 58-75 Emperor Cheung... 145 

Emperor Chang 76-88 Emperor Chih 146 

Emperor Ho 89-105 Emperor Huan 147-167 

Emperor Shang 106 Emperor Ling 168-189 

Emperor An 107-125 Emperor Hsien 190-219 



Chapter I 



INTRODUCTION 

C>00<X>0<c><><><><><><><^^ 

There is some danger attached to the use of the word 
"vignette" in the title of this book. "Vignette" has long 
been used to denote those little floral patternings that, nota- 
ably in nineteenth-century publications, fill up the blank 
spaces on the opening and closing pages of the chapters. 
They can be pleasing enough to the eye, but in so many cases 
they have nothing to do with the meaning of the book. They 
are, therefore, in the full sense of the modern opprobious 
term "extraneous ornament." "Vignette" has, however, of 
late years acquired more reputable nuances of meaning, as, 
for example, a photographic vignette in which the attention 
is focussed on the subject in a characteristic pose and the 
non-characteristic outlying parts of the figure are deliberately 
blurred. Since the main part of this book is to present Pan 
Ku (a.d. 32-92) and Chang Heng (a.d. 78-139) as they 
appear in their respective two prose-poems on the Two Capi- 
tals,^ "vignette" in this sense would seem appropriate. 

Further, in the eighteen-eighties George Saintsbury wrote 
"to finish off and vignette isolated sketches of manner, char- 
acter, and thought with more precision . . . than is possible 
and suitable in prose" (see Oxford Dictionary^ ad loc). That 
usage also validates the term in connection with this book, 
written primarily for Western readers. 

For the benefit of the general reader who may be un- 
familiar with the early history of China, it is perhaps desira- 

^ An annotated French translation of the two prose-poems by Pan Ku 
may be found in G. Margoulies Le "Fou^^ dans le Wen-siuaUy Paris, 1926, 
pp. 31-74, and a German translation of the two by Chang- Heng in E. von 
Zach, Die Chinesischen Anthologity Ubersetzungen aus dem Wen Hsuan^ 
ed. I. M. Fang, Harvard- Yenching Institute Studies xviii, 1958, vol. I, 
pp. 1-37. The von Zach translation was first published in Sinologische Beit- 
rage 2 (1936). 

[ 3 ] 



INTRODUCTION 

ble at this point to supplement the chronological chart at 
the beginning of this volume with a brief outline of the his- 
torical developments which led up to the creation of the 
Later Han society in which Pan Ku and Chang Heng, the 
two writers with whom this book is chiefly concerned, both 
lived. 

The immensely long period known as the Chou Dynasty 
began with the conquest of the Shang kings in Honan by pow- 
erful vassal chieftains from the west. The traditional date of 
conquest, 1122 b.c, is probably about a century too early. 
The Chou kings ruled in Shensi and exercised some sort of 
control over a large number of small vassal states scattered 
over the whole of northern China. The nobles who governed 
these states were generally related to the Chou kings either 
by blood or marriage. In 771 b.c. the Chou capital was sacked 
and the Chou king slain by barbarian invaders, and a rem- 
nant of the royal house set up a new capital farther east in 
LoYANG. From this circumstance the era ending in 771 b.c. 
is known as Western Chou and the era following the re- 
moval of the capital Eastern Chou. It will be seen that 
there is a parallel in the Han dynasty when the capital was 
moved from Ch^ang-an to Loyang in a.d. 24. 

After the transfer of the capital, the control of the Chou 
kings over their vassals became more and more ineffectual. 
The vassal princes enlarged their states by opening up new 
lands, by the elimination of the "barbarian" peoples who had 
formerly lived side by side with the Chou settlements, and 
by the conquest of weaker rivals. But since no single state 
was strong enough to withstand unaided the onslaught of 
powerful external enemies like the barbarians of the north 
or the peoples of the Huai and Yangtze river valleys, a suc- 
cession of Hegemons arose — rulers of outstandingly rich and 
powerful states who, acting nominally on behalf of the effete 
Chou kings, led confederacies of the Chou states in war and 
exacted tribute from them. This period (722-481 b.c.) is 
called the Spring ^nd Autumn period after the name of the 

[ 4 ] 



INTRODUCTION 

chronicles of the state of Lu which are our chief source for 
its history. 

The last two centuries of the Chou dynasty, from 403 b.c. 
to the end of the dynasty, are known as the Warring States 
period. In this era the Chou kings ceased to exercise even a 
nominal control over the other states. The rulers of the 
seven most powerful states called themselves kings and each 
aspired to unify China under his own sway. They ruled over 
territories as large as European countries with large popu- 
lations which economic, technological, and cultural advances 
had raised to a level of civilisation vastly superior to that 
of Chou society in the early days of the dynasty. These great 
states were almost continuously engaged in war and diplo- 
matic intrigue against each other. Finally the powerful west- 
ern state of Ch4n eliminated all the other states, and in 221 
B.C. the king of Ch^in became emperor of a unified China 
and founded the Ch'in Dynasty. He abolished the feudal 
nobility, set up a great bureaucracy, standardised weights and 
measures and the Chinese script, built roads, canals, and the 
Great Wall of China, and extended his conquests into Korea 
and Indo-China. His dynasty did not long survive him, how- 
ever, since these immense achievements were based on a ruth- 
less and tyrannical oppression of the people, who quickly 
disposed of his incompetent and weakly successors. Of the 
various adventurers who strove for control in the ensuing 
anarchy, a soldier of humble origin called Liu Pang (later 
known as Kao Tsu, the "August Founder" of the Han dy- 
nasty) eventually succeeded in establishing a new dynasty. 
This Han Dynasty had its capital at Ch'ang-an," near the 
modern Sian in Shensi. 

Of the emperors who succeeded Kao Tsu, the most re- 
markable was the Emperor Wu (140-87 b.c), both for the 

^ The spelling adopted here and elsewhere for the name of this city is 
in accordance with the Wade-Giles system of romanisation. The name is 
no longer current, having been replaced, in modern times, by Sian, In the 
case of the other capital, Loyang, the name still exists today and so the 
conventional Western spelling has been employed. 

[ 5 ] 



INTRODUCTION 

length of his reign and for the remarkable expansion of 
Chinese power and influence in Asia which took place dur- 
ing it. In the steppes north of China a nomad empire of 
the Huns had come into being at about the same time as 
the Ch'in emperor's establishment of a unified empire in 
China. The exploits of Emperor Wu's armies drove back 
the Huns and ended for a time the hitherto constant threat 
of their incursions into Chinese territory. Chinese armies 
penetrated far into Central Asia, and the ChSn conquests 
in Korea and in the south were renewed. 

The splendours of Emperor Wu's reign resulted in con- 
siderable impoverishment of the economy, and the emperors 
following him were mostly incompetent. In a.d. 9 a member 
of the Empress' clan called Wang Mang overthrew the dy- 
nasty and usurped the imperial title. Wang Mang's ill-judged 
reforms of the economy led to great confusion and distress 
resulting in a great peasant uprising called the Red Eyebrows 
rebellion. This was closely followed by further risings led 
by various Han nobles. One of them, Liu Hsiu, eventually 
prevailed and became Emperor of a restored Han dynasty 
in A.D. 23. Wang Mang was killed and the imperial palaces 
burnt to the ground when Liu Hsiu's soldiery sacked Ch'ang- 
an. Millions are said to have perished in the fighting and 
massacres which took place during these disorders. This re- 
stored Han dynasty is usually called Later Han to distin- 
guish it from the Former Han dynasty deposed by Wang 
Mang. The capital was removed to Loyang. The first half- 
century of the Later Han period was a time of economic re- 
covery and expansion. This was the period in which the first 
of the two authors with whom we are concerned lived. From 
about A.D. 80, however, there was a steady deterioration in 
the internal political situation, notably in the struggles be- 
tween the powerful eunuch party at court and the scholar 
bureaucrats in the provinces, which ultimately led to the 
dynasty's overthrow. 

[ 6 ] 



INTRODUCTION 

Apart perhaps from one or two Western historians of the 
calibre of Chavannes and Henri Maspero, the distinctive 
features of the Later Han regime about a.d. ioo are for the 
most part blurred and haphazardly distinguishable from those 
of a century earlier when Former Han was on its last legs. 
Yet it is common knowledge today that in the half-dozen 
generations from lOO b.c. to a.d. ioo something of extraor- 
dinary moment happened to the ideologically inchoate peo- 
ples who inhabited the vast continental region east of the 
Himalayas. That "something" has been traditionally denoted 
as the triumph of the Ju Tao, the doctrines upheld by the 
main bulk of the literates of that age. Recently the develop- 
ment of a more critical study of the records has led to the 
coining of a new term among Sinologists, "State Confucian- 
ism." The term is useful, as far as it goes, but it is submitted 
here that there is still a considerable blur in Sinologists' minds 
as to what actually happened. 

Since those half-dozen generations were, in the fullest sense 
of the term, epoch-making in the history of the Chinese 
people, there is crying need for vignetting "sketches of man- 
ner, character, and thought," and for doing this on the basis 
of documents identifiable as to date and authorship. For this 
reason I came to a somewhat dangerous decision. Instead 
of giving complete translations of Pan Ku's and Chang Heng's 
respective prose-poems on Ch^ang-an, the old capital, and 
Loyang, the new capital, I summarized those parts of the 
texts where it seemed least damage would be done to the 
full explication of the authors' minds. (See Chapters III-VI. 
The indented material is direct translation j the remainder is 
summary and paraphrase.) Thus room was made within the 
bounds of one volume for four considerable chapters of cri- 
tique (Chapters VH-X). In these the raw materials for vi- 
gnettes of characteristic manners and modes of thought have 
been set down just as they emerged to view when the senti- 
ments expressed in the poems were related to such other 
documents as appeared relevant. In the last chapter, after a 

[ 7 ] 



INTRODUCTION 

final scrutiny of Pan Ku's and Chang Heng's reliability as 
contemporary witnesses, the more important of these vignettes- 
in-the-rough have been examined under systematic headings. 

The quotation from Saintsbury contained these words: 
"sketches . . . with more precision . . . than is possible and 
suitable in prose." A provocative sentiment, calculated to 
surprise and even to arouse dissent from the research type of 
historian. Now, whatever may have been in Saintsbury's mind 
when he wrote those words, the reference to poetry as being 
more precise than prose can and should be is highly relevant 
to the subject matter in Chapters III to XL The whole of 
this book centers round the four poems : very lengthy poems, 
containing prose elements but nonetheless highly poetical 
with their rhymes and their studied patterns of rhythmic 
sentence structure. In a word they are descriptive poems in 
which the authors' dominant aim was to depict certain ob- 
jects {wu) of contemporary interest, not to explore the rami- 
fications of their own emotions. Yet emotion comes in from 
time to time. 

To the would-be reader in search of facts this feature of 
the documents under scrutiny may damn the book right away. 
Yet the rigorist historian, before turning his back on the 
evidence to be found in these poems, might do well to pause 
and consider three indisputable facts : ( i ) The two authors, Pan 
Ku and Chang Heng, were noted men in their day, their births 
and deaths are on record, as also considerable information 
about their careers. (2) Pan Ku was a historian by training 
and profession, of the highest repute in the annals of Chinese 
historiography j Chang Heng was a mathematically-minded 
astronomer, eminent as a trail-blazer in the scientific study 
of the heavens. (3) Both men wrote much in prose, but they 
elected to embody their respective depictions of the Two 
Capitals and the accompanying regimes in that new fu genre^ 

^ The fu^ a literary medium which became very popular during- the Han 
dynasty, was a sort of poetic essay, generally descriptive, in which prose 
and verse were mixed. Its most characteristic features are the consistent 
use of parallelism and a highly ornate vocabulary. 

[ 8 ] 



INTRODUCTION 

to which the genius of Ssu-ma Hsiang-ju had given so shaq) 
a dialectical edge. 

Since external and internal evidence demonstrate that the 
two authors took immense pains to make their compositions 
effective appeals to their age, it would seem rather hazardous 
to assume that the precisional quality which characterized 
their professional studies should not inform their poetical 
efforts. Each had an axe to grind — possibly a personal one, 
possibly a nobly patriotic one — but since when did research 
historians refrain from examining documents because they 
showed signs of prejudice and tendenz? Does not the scien- 
tific approach to history take it as inevitable that every writer 
on his own period, even the recorder of statistics, has some 
particular conscious or unconscious slant to his mind? 

The emphasis on these three facts has, it is hoped, given 
assurance on the evidential value of the four documents pro- 
posed for examination. There is a fourth consideration arising 
out of the epistemological axiom that rational knowledge 
comes by comparison, and the more clearly the compared 
objects can be envisaged the more precise and more reliable 
the results in knowledge. In the case under consideration 
the first object was the capital, Ch^ang-an, which had been 
looted and largely burnt some sixty years before Pan Ku 
wrote, and the Former Han regime which expired twenty- 
one years before that looting and burning. The second con- 
trasted object was Loyang, capital of the Later Han regime 
of which Pan Ku was a court official. Some years after Pan 
Ku died, Chang Heng, having read Pan Ku's detailed de- 
scriptions of the Two Capitals, wrote a considerably longer 
comparison of them. Both Pan Ku and Chang Heng had 
each his own angle of vision, made his own particular em- 
phases, but had his own inner compulsion to state his definite 
impressions. The results, the four poems, reveal agreements 
and disagreements. At numberless points the narratives give 
details which supplement the information given in one or 
the other document. The reflections on the varied phenomena 

[ 9 ] 



INTRODUCTION 

treated are often strikingly individual. Comparison, however, 
cannot be completely fari fassu, for Chang Heng was cog- 
nizant of events which occurred between a.d. 89, the last 
possible date for the completion of Pan Ku's poems, and a.d. 
126, the year in which Chang Heng presented his poems to 
the throne. (See Chapter II for a discussion on the two dates.) 

The conclusion stares us in the face that these four poems 
might well contain a considerable store of information. What 
is more, the store is a unique one, since nowhere else in the 
literature of the Han era can we find two such near-contem- 
porary authors dealing with the same subjects of discourse in 
so intimately comparable a fashion. As to the relative intelli- 
gence and rational acumen of the two authors, that will be 
one line of enquiry in the critiques on the poems. But it is 
as well to make clear in an introductory chapter that their 
epistemological consciousness was not so naive as it is tempting 
to suppose with men belonging to so superstitious an age as 
that of Later Han. For example, in the language current in 
their day there was the term '^hslan^^ (representational 
image) standing in contrast to a "w^" (a concrete object of 
sense perception). Behind this distinction lay the recognition 
of the part played by the five senses in creating mental images. 
There was also the recognition that knowledge derived from 
hearsay had less veridical value than knowledge gained by 
direct observation. The poems reveal that Pan Ku and Chang 
Heng were mindful of this — although whether they and their 
contemporaries were consistent in their application of the dis- 
tinction is open to question. 

With regard to such scaffoldings to thought on concrete 
matters, no attempt is made here to list them in relation to 
the mid-Han literates: helpful as Fung's History of Chinese 
Philosofhy and Bodde's translation of it are,* the necessary 
data are still far from being classifiable on reliable semantic 

* Fung Yu-lan, Chung-kuo che-hsueh shih (2 vols.), Commercial Press, 
Shanghai, 1946. Fung Yu-lan, A History of Chinese Philosophy ^ Vols. I 
& II, tr. D. Bodde, Princeton University Press, 1952-53. 

[ 10 1 



INTRODUCTION 

foundations. But they are of the utmost importance, and it 
is hoped that this present study of two outstanding mid-Han 
minds may throw some light on the problems involved. The 
chief of these problems is, as all Sinologists are so painfully 
aware, to date with anything approaching certainty or preci- 
sion the Five so-called "Confucian" Classics.^ They constitute 
the largest and most important, although by no means the 
only important, source of ideological information for the 
period. Historians can agree that their existence in recogniz- 
able shape is guaranteed by the listing of them in Liu Hsin's 
"Catalogue"^ of the date a.d. i or thereabouts. But the same 
passage reveals that there were variant texts of these Scrip- 
tures in the imperial library, whilst the researches of the 
Ch'ing scholars and modern critics have driven home the 
fact that for these "Five Scriptures" (with the possible ex- 
ception of the "Spring and Autumn Annals") there is no 
conclusive evidence as to who the authors and compilers were, 
and how few or how many redactors and amplifiers had a 
hand in bringing these texts to the uncorrelated condition 
disclosed by Liu Hsin's "Catalogue." Since the inception of 
these documents dates far back in the Chou Dynasty, the 
vast bulk of the statements in them is only hypothetically 
dateable. Yet their influence in mid-Han times is writ large 
in every phase of thought and social observance, so that the 
historian today cannot afford to ignore them. When he tackles 
this problem, he finds himself in the situation of Gilbert 
and Sullivan's billiard sharper, "playing extravagant matches 
with a twisted cue on a cloth untrue and elliptical billiard 
balls." 

From the point of view of the ideological historian here 
is perhaps the paramount incentive to the student to break 

M.e. (i) the "Changes" (/ Ching), (2) "History" {Shu Ching) , (3) 
"Odes" \Shih Ching), (4) "Rituals" (Li Chi, etc.), and (5) "Spring and 
Autumn Annals" {Ch^im ChHti). In this book these five are collectively re- 
ferred to as the Scriptures. 

^The "Monograph on Bibliography" (/ Wen Chih) in Pan Ku's "His- 
tory of the Former Han Dynasty" {Han Shu) . The monograph was mostly 
written by Liu Hsin. 

[ II ] 



INTRODUCTION 

out of that charmed Scripture circle and be content for a 
time to concentrate on dateable documents by well-known 
authors late in that key mid-Han age. In their four poems 
Pan Ku and Chang Heng make one or two references to "the 
Scriptures" in general. They do so in the respectful language 
one would expect from members of the Later Han court, 
but their outlook on life is by no means restricted to guidance 
from Scripture texts. Pan Ku, indeed, expresses his disgust 
at the benighted practice of proving by proof-texts. On the 
other hand, there is the fact that throughout the 2360 verses 
which make up the four poems^ the phrasing is constantly 
remindful of Scripture sentences.^ This, of course, is what 
one would expect from men whose education consisted mainly 
in learning to read and understand the Scriptures. But the 
research mind can hardly be content to be stopped by that 
reflection. New light on meaning comes from seeing those 
phrases caught out of their Scripture contexts, adapted to the 
poet's use in metrically formed sentences, sometimes in ways 
which might stagger the original authors. 

The profit, however, goes beyond that. The vivid person- 
ality of each author is constantly in evidence, and by his 
mastery of language we are able to see things through his 
eyes: emperors and their palaces and entourages, their duties 
and amusements, as also the busy life of the city streets, the 
merchants and pedlars, the street-corner arguers, the gang- 
sters in the suburbs lying in wait for the unwary, and the chil- 
dren in procession expelling the demons of disease. At the 
very least the people of that far-back age become more nearly 
alive as human beings, cease in a measure to be lay figures 
in the mass, vague representational images created in the 

'^ A small margin of error must be allowed for in this assessment. With 
Pan Ku's Preface (304 characters), the total number of characters comes 
at a roug-h estimate to well over 12,400. 

^ The seventh-century commentary on Wen Hsiian by Li Shan consists 
mostly of illustration of the text by means of citations from the Scriptures 
and other works. In the case of Chang Heng's "Fm on the Eastern Capital" 
many of the citations are taken from Chou Li^ suggesting that Chang Heng 
may have been familiar with that work. 

[ 12 ] 



INTRODUCTION 

modern's unconscious imagination. There is even more than 
that: dateable descriptions of imperial sacrifices, of a great 
durbar for foreign potentates and domestic princes and nobles, 
pictures in concrete detail of the throne as it functioned in 
Later Han. Also here are examples of what could be said 
to Later Han ancestor-worshipping emperors in criticism of 
earlier occupants of their dynastic throne. 

With regard to those Scripture parallels, it is urged that 
the linguistic researcher would do well to tread warily. He 
dare not blindly assume that in all cases here is evidence 
guaranteeing that such and such a passage was in such and 
such a Scripture at such and such a point. It is quite possible 
that a certain phrase might have come into quite recent cir- 
culation, a newly coined idea, outcome of the intellectual 
eclair els sement which characterized the mid-Han age, and 
then either by addition of some amplification to a Scripture 
or by way of being an elucidatory redaction have crept into 
the sacred text. Three considerations demonstrate the need 
for caution. One is that although after Emperor Wu's reign 
(140-87 B.C.) it may have been difficult for a group of offi- 
cial expositors to insert additions, yet Liu Hsiang found the 
tablets and scrolls in the imperial library imperfect and out 
of their proper order j and we do not know what he and his 
colleagues did to make order out of disorder and sense out 
of non-sense. The second is that during the two decades of 
the Wang Mang regime (a.d. 9-22) the propagandist mind 
was strong in court circles, and also that whilst Kuang Wu 
(a.d. 25-57) was desperately engaged in establishing his su- 
premacy, his Erudites were both guardians of the sacred texts 
and propagandists for their master's cause. Thirdly, the Ts'ai 
Yung inscription of some sort of textus recepus^ on pillars 
of stone in a.d. 175 was sanctioned because there was doubt 
as to what the authentic texts did or did not contain. In Chap- 

^ Now known only from fragmentary references in the writings of later 
scholars. 

[ 13 ] 



INTRODUCTION 

ters VII-X this state of affairs crops up here and there for 
consideration in passing. In Chapter XI is an attempt to 
draw the earlier suggestions together and clarify their sig- 
nificance, but no attempt is made to do more than set down 
one student's reflections. 

I make no apology for the microscopic attention to detail. 
As the late R. G. Collingwood, a brilliant archaeologist and 
student of history as well as an Oxford professor of meta- 
physics, maintained in his inaugural lecture, the study of his- 
tory is in the last resort a study of the individual and con- 
crete.^^ To this I would add this amplification : those fractional 
yet at times highly revealing concrete bits of information 
which have survived the wrack of time till now. The only 
question with regard to a particular culture and period of 
history is whether it is sufficiently in the main stream of world 
history to warrant the labor of envisaging. From that angle 
of approach, in this "Year of Grace" 1956, a quite impressive 
case may be made out for the history of the Han era. To 
take only the history of Western civilization for comparison, 
the Greek ideological and institutional achievements were 
followed by the Romany and it was Rome which first en- 
visaged the prerogatives and duties of world-dominion and 
started to integrate the various races under its influence. The 
power of Rome crashed eventually, but not before a new 
mysteriously energising force, known as "Christianity," had 
become a re-integrating influence from which sprang a new 
distinctive Western culture which throughout its succeeding 
phases has more and more become relevant to other races 
besides those of Europe. If "Rome" had not succeeded 

^^ R. G. Collingwood, The Historical Imagination (Clarendon Press, 
1935) • E.g. p. 5: "No doubt, historical thought is in one way like percep- 
tion. Each has for its proper object something individual. What I perceive 
is this room, this table, this paper. What the historian thinks about is Eliza- 
beth or Marlborough, the Peloponnesian War or the policy of Ferdinand 
and Elizabeth. But what we perceive is always the this, the here, the 
now ... the things about which the historian reasons are not abstract but 
concrete, not universal but individual, not indifferent to space and time 
but having a where and a when of their own." See also CoUingwood's 
The Idea of History (Clarendon Press, 1946), a posthumous production 
of his uncompleted work. 

[ 14 ] 



INTRODUCTION 

"Greece" and "Christendom" not succeeded "Rome," what 
then? 

The "China" of the Han era, with its envisagement of the 
prerogatives and duties of world-dominion, constitutes a Far 
Eastern phenomenon comparable to Rome, particularly as the 
Han alertness to philosophical values stemmed from the Chou 
thinkers, as Rome did from Greece. Pan Ku optimistically, 
Chang Heng less optimistically, believed that the civilisation of 
Great Han would endure for a thousand ages. In that belief 
Pan Ku was mistaken and Later Han crashed a hundred years 
after Chang Heng's death. Yet after five centuries of almost 
continual disruption and barbarian infiltration a unified China 
of more than Han integral dimensions re-emerged, to be fol- 
lowed by periodic effectual assertions of cultural cohesion and 
civilizing power. The Chinese of these later ages have had 
a synonym for their cultural unity, "sons of Han." 7/ "Great 
Han" had not supervened on the finally disrupted Chou or- 
der, and if what "Great Han" achieved in cultural integra- 
tion had not been strong enough to survive the strains and 
stresses of the following five centuries, to be followed by 
"T'ang China" and "Sung China" . . . , what then? The "ifs 
of history" when categorically set forth admit of no categori- 
cal answer. They remain question marks and very dangerous 
ones. But as such they are useful in driving the historian to 
pursue his dedicated task of ascertaining such concrete facts 
as may illumine the interplay of the great cultural forces 
which in their contemporary forms are today shaping the 
course of world events. That "Great China" is one of those 
forces is self-evident, as much as "Great America," "Great 
Britain." In Henry Steele Commager's impressive The Amer- 
ican Mind}'^ there stands in the forefront Santayana's pregnant 
dictum, "to be an American is of itself almost a moral con- 
dition, an education and a career." To Pan Ku and Chang 
Heng at the turn of the first century a.d., to be a Han Chi- 
nese had very much the same portentous meaning. 

^^ H. S. Commager, The American Mind, An Interpretation of American 
Thought and Character since the i88o^s, London, 1950. 

[ 15 ] 



Chapter II 



BIOGRAPHICAL DETAILS 

NOTE: The following two accounts are constructed mainly from 
information given in the biographies of Pan Ku and Chang Heng, 
chs. 70 and 89, respectively, in the "History of the Later Han 
Dynasty" (Hou Han Shu), the biographical section of which 
was composed by Fan Yeh (398-446) of the (Liu) Sung Dy- 
nasty. Pan Ku's story has been supplemented with information 
from the biographies of his father, Pan Piao, his brother. Pan 
Ch'ao, and his sister, Pan Chao; also from his autobiographical 
preface (Hsu Chuan), the last chapter of the "History of the For- 
mer Han Dynasty" (Han Shu). A few details come from other 
parts of the 'Tlistory of the Later Han Dynasty" in which ref- 
erences to Pan Ku occur. Chang Heng's story is almost entirely 
from his biography, but certain details are derived from modern 
histories of mathematics, notably that by Professor Li Yen (Chung- 
kuo Ku-tai Shu-hsiXeh Shih-liao, Chung-kuo K'o-hsiieh T'u-shu 
I-ch*i Kung-ssu, Shanghai, 1954). 

Fan Yeh depended on records in the archives at Chien-yeh 
(Nanking), the Chin capital, amongst them the Tung Kuan Han 
Chi, which was compiled more or less contemporaneously by im- 
perial orders. How much of this document was in existence in the 
fifth century is in question, but some of it is still extant. Fan Yeh 
died before his projected work was completed, and the Chth 
(monographs) he had in mind to write were later supplied by 
the commentator Liu Chao (flor. 502-519) from the Hsu Han 
Shu of Ssu-ma Piao (240-306) and incorporated in Fan Yeh's 
work. Today, therefore, we are unable to estimate what Fan Yeh's 
final word would have been, e.g. in regard to the biographies he 
compiled. But in a letter he wrote to his family just before his 
execution (printed in Wang Hsien-ch^ien, Hou Han Shu Cht- 
chieh, Introduction) he referred to his aims in compiling them; 
and in the light of that highly illuminating document — a landmark 
in Chinese historiographical acumen — his biographies of Pan Ku 
and Chang Heng are assumed here to be generally trustworthy. 

[ 16 ] 



BIOGRAPHICAL DETAILS 

Pan Ku {tzu Meng-chien), born a.d. 32, died a.d. 92, came 
of a family of considerable distinction in the service of the 
state. Its fame in this respect dated from the last decades 
of the Former Han regime, but the family tradition was that 
they were descended from the royal house of Ch^u State. In 
the Ch4n First Emperor's time an ancestor packed up and 
went off to the frontier lands in the northwest where he made 
money in sheep and cattle. Two of Pan Ku's great-uncles had 
careers at court, and one of them. Pan Yu, was ennobled and 
finally became collaborator with Liu Hsiang in the task of 
putting the imperial library in order. The Emperor presented 
him with books from his store, thus adding to the family 
library which scholars came from far to see. One member 
of the family was an authority on Taoist books. At that time 
the Pans were affluent, although whether they continued to 
be so is doubtful. Pan Yu "died untimely" when Wang Mang 
began his usurpation. Pan Piao, Pan Ku's father, remained 
loyal to the Han house and retired into the west. Being 
badgered by the war-lord where he was to assist in his rebel- 
lious plans, Piao went elsewhere and took service with a loyal 
general. When Emperor Kuang Wu came to power. Pan Piao 
was appointed Governor of Hsu (North Kiangsu), but after 
a time was released from office owing to ill health. Later he 
was summoned to court and held a succession of posts, all 
of them a source of discomfort to him. He did not achieve 
the distinction of ennoblement. At the age of fifty-two he 
died while magistrate of a district. "His officers and people 
loved him." 

The upper-class way at that time was for lads of promise 
to be sent for schooling in the capital. Pan Ku, although 
clearly blest with brains, was for some unknown reason 
trained at home. His father cherished a scheme for writing 
a history of the Former Han dynasty, and it was in this 
austerely studious atmosphere that Pan Ku developed his 
intellectual powers. When his father died he continued with 

[ 17 ] 



BIOGRAPHICAL DETAILS 

the history- writing project. Before long he ran into trouble. 
A report reached the court that a private person was engaged 
in writing a history — a punishable offence. His brother, Pan 
Ch'ao, was in Loyang at the time, being engaged in some 
minor copyist capacity. Pan Ch'ao immediately took steps to 
protect his brother's name, and the result was that Pan Ku 
was summoned for interrogation by Emperor Ming (a.d. 58- 
75). The Emperor not only decided in his favor, but also 
gave him access to the palace archives. 

Thus the prospect of a successful court career opened before 
Pan Ku's eyes. Yet, although he was near to his fortieth year, 
his official standing was only that of a lang (a cadet awaiting 
appointment), his allowance a mere few hundred piculs of 
grain a year.^ His brother also was finding no scope for his 
abilities in the capital. For Pan Ku the only course was to 
put on all speed and complete his Han history. There is no 
external evidence by which to date his beginning of the "F« 
on the Two Capitals," but it seems probable that the theme 
began to take shape in his mind during his first impressionable 
years at Emperor Ming's court. ^ Some time in Emperor 
Chang's first reign-period (76-83) the task of over twenty 
years was completed and the "History of the Former Han 
Dynasty" was submitted to the throne, all, that is, except 
the monograph on T^ien Wen (astronomical observations).^ 
It was accepted with approval and among the scholars re- 
ceived with enthusiastic welcome. Yet no well-paid post came 
to him. All that the Emperor did was to employ Pan Ku in 

^ From Chia K'uei's biography {Hou Han Shu^ ch. 66) it appears that 
Pan Ku was associated with the intimates of the imperial circle, and he 
may have had a hand in preparing the "Acts of Emperor Kuang- Wu" (see 
Tung Han Hut Yao^ ch. 12). Conceivably his apparent failure to obtain 
any substantial sig^n of imperial approval was due to the jealousy of Chia 
K'uei. 

^ Cf. Pan Ku's Tien Yin (Wen Hsilan^ ch. 48) written in collaboration 
with Chia K'uei and others at Emperor Ming's instigation. Some of the 
emphases characteristic of his fu are found there in an undeveloped form. 

^ Han ShUy T^ien Wen Chih^ completed by his sister, Pan Chao, after 
Pan Ku's death. 

[ 18 ] 



BIOGRAPHICAL DETAILS 

writing a resume of the findings of the great scholars' con- 
ference in the White Tiger Hall. 

Pan Ku then set to work on his "F^ on the Two Capitals." 
In what year he finished and presented them to the throne 
is doubtful, but the immediate background to his meditations 
must have been that of Emperor Chang's court.* Again there 
was universal admiration, but still no higher post was given 
to him. Since the record of his character has it that he was 
"amiable and accommodating, not exalting himself above 
others because of his talents," he can hardly have been com- 
fortable in mind. Finally General Tou, brother to the Em- 
peror's chief consort, went on campaign and took Pan Ku 
as a member of his staff. Again fortune was cruel. On Tou's 
return the animosity against him at court came to a head. 
He was impeached, and Pan Ku was accused along with him. 
Before that blow came, an unruly slave of Pan Ku's household 
had offended the Intendant of the Capital. He bided his time, 
and then, when Pan Ku's name was under a cloud, he dis- 
patched lictors to waylay him and throw him into prison. 
Before his friends could come to his rescue, he died. The 
record has it that the Intendant was later "reprimanded" and 
one of his officers "punished." 

Pan Ku's "History of the Former Han Dynasty" {Han 
Shu) in a hundred chapters has made his name a household 
word among scholars all down the agesj and today, in spite 
of the uninhibited approaches of modern critical scholarship, 
his reputation stands as high as it ever did. As an overall 
view of a period his book is an astonishingly mature exhibi- 

* Li Shan (d. 689), the first great commentator on the two fu^ and a 
man whose cautious statements created a general confidence in him, re- 
marks at the outset of his commentary that Pan Ku presented his /// "with 
the intention of remonstrance, and Emperor Ho was greatly pleased." Since 
Emperor Ho was only thirteen when he succeeded in 89 and his entourage 
was not one likely to welcome remonstrance, this seems unlikely, particu- 
larly as Pan Ku died in 92. In any case, since Emperor Chang died in his 
thirties, Pan Ku must have written the poems with the intention of pre- 
senting them to him. 

[ 19 ] 



BIOGRAPHICAL DETAILS 

tion of the science and art of historiography. To conceive 
of a period history as entailing treatment from four differ- 
ent angles of approach, each shedding a complementary light 
on what happened, and to do this not in rough outline but 
in detail on judiciously handled information derived from 
state documents, may be described as an intellectual achieve- 
ment which has no counterpart in other ancient cultures/ 
The plan of the book owes much to Ssu-ma Ch4en's "Record 
of History" {Shih Chi) of a century and a half earlier, but 
Pan Ku greatly improved on the patterns embodied in that 
work. One question ranges round the extent to which Pan 
Piao conceived the scheme in its entirety and whether he 
was the author of parts of it.® The question has to be asked, 
since Pan Ku has from time to time been accused of being 
unfilial in taking all the glory to himself. It seems un- 
likely, if Pan Ku had before his eyes the example of 
filial piety set by Ssu-ma Ch^en, that he would have de- 
liberately omitted all recognition of his father's claim to a 
substantial share in the final product. 

Pan Ku has also a great name among connoisseurs of literary 

^ To enable Western historians to get some idea of this achievement in 
historiographical method, here is an abstract of the "History of Former 
Han Dynasty" {Han Shu) : Section / (12 chapters) : a record year by year 
of the main acts of the twelve emperors with some account of the circum- 
stances in which the more important decisions were made. Section II {Z 
chapters) : tables containing lists of the royal princes, the ranks of nobles, 
the officials of each reign, and the great men (graded in 9 classes) of past 
ages. Section III (10 chapters): substantial monographs on the Principles 
of the Calendar, Rites and Music, Principles and Applications of the Law, 
Food and Commodities, Sacrifices in the Suburbs and Outlying Localities, 
Operations of the Five Forces in Nature, Geography (Irrigation Systems 
etc.), Catalogue of the Imperial Library. Section IV (170 chapters) : biog- 
raphies or short memoirs of 240 royal princes and all other kinds of emi- 
nent men of the dynasty, including the most eminent litterateurs. This 
section also includes chapters on "the Scholars' Grove" (with details about 
27 famous teachers), on Lenient Officers (6), on Tyrannical Officers (12), 
on the Wealth of Merchants (11), on Wandering Bravoes (7), on Palace 
Favorites (7), on the surrounding Barbarian Tribes, on the Empresses, on 
Wang Mang the usurper, on various aspects of the Pan family, develop- 
ments of scholarship, and the plan of the book as a whole. 

® See H. H. Dubs, History of the Former Han Dynasty^ vols, i & 2 
(Baltimore, 1938, 1944), passim. 

1 20 ] 



BIOGRAPHICAL DETAILS 

Style. According to Fan Yeh, "forty-one of [Ku's] pieces were 
extant in his day." Of these the ones treasured in the Wen 
HsilaW show competency in many of the polished literary 
modes practised in Later Han, including jesting conversa- 
tions at feasts. But his outstanding poetic achievement was 
the ^^Fu on the Two Capitals." There he followed Ssu-ma 
Hsiang-ju's (i 79-1 17 b.c.) original lead in transmuting the 
ju genre into a depiction of things instead of emotions, of 
large-scale objects of attention in contrast with mythological 
or trifling ones. In sheer length Pan Ku's "Fw on the Two 
Capitals" far surpassed any previous ]u composition. It is 
also distinguished by its interplay between two dramatis fer- 
sonae — a device used by Hsiang-ju to give point to his irony 
and wit. Of the Pan Ku compositions preserved in the Wen 
Hsuany his Tien Yin (ch. 48) seems to me the more revealing 
document, both in its foreword and in the eulogy itself. It 
brings out most vividly Pan Ku's feeling of dependence on 
Emperor Ming's personal favor. 

Chang Heng {tzu P^ing-tzu), born in 78, died in 139, came 
of a provincial family in the south of the modern Honan. 
His grandfather was Governor of the Ch^eng-tu Comman- 
dery (Szechuan) during the last two years of Former Han, 
but apart from him there is no record of the family being 
distinguished in any way. It must have had means because 
Chang Heng was sent to the capital for schooling. He grad- 
uated in due course, being recommended as a hsiao lien (lit. 
"filial and upright"). Instead of seeking an official career, 
he went home and devoted his time to learning. He did not 
find common intercourse easy. Mathematics in relation to the 
heavens was his forte. In the course of time he was able to 
produce a new armillary sphere, and to construct a chart of 
the sky which was both detailed and clear. Philosophically 
he had a passion for Yang Hsiung's (53 b.c.-a.d. 18) Hsilan 

"^ Famous antholog-y of verse and prose compiled by Hsiao T'ung in the 
6th century. 

[ 21 ] 



BIOGRAPHICAL DETAILS 

Ching which he placed on a level with the "Five Scriptures." 
His technological ability showed itself in various ways. For 
instance, he made a flying machine — which apparently could 
raise itself from the ground but did not stay long in the air. 
In 132 he produced a seismographic apparatus, able to record 
the direction of earth tremors which were imperceptible to 
the ordinary senses. 

It was in those years of seclusion that he was stirred to 
emulate Pan Ku by writing a "Fw on the Two Capitals." 
Ten years had passed before he finished itj it is not known 
when he presented it to the throne. It is difficult to believe 
that he did so at a time when an official career in keeping 
with his talents was opening before him. This was at some 
unspecified time in Emperor An's reign (a.d. 107-125), pre- 
sumably when An was twenty or more years old. He sent 
for Chang Heng and after making him a cadet-in-waiting 
appointed him Director of Recording in charge of the sub- 
department engaged in recording astronomical and meteor- 
ological events. About the time of young Emperor An's death 
the appointment apparently lapsed,^ and it seems likely that 
it was at this time that Chang Heng completed his "F« on 
the Two Capitals," presenting it to Emperor Shun (a.d. 126- 
144).^ Emperor Shun "again transferred him," and he was 
reinstated as Director of Recording. Finding himself out of 
sympathy with the dominant officials, Chang Heng resigned 

® There is no record of when and why, nor of whether or not he retired 
to his home, 

^ Chang Heng- could hardly have passed for government service until 
his early twenties. On the basis of the biography's "ten years" spent in 
composing his "F« on the Two Capitals," this brings the date of its com- 
pletion down to the middle years of Emperor An's reign, i.e. about a.d. 120. 
It was about then that a carriage was sent to bring him to the capital. 
Dowager-Empress Teng died in 121, and it is difficult to believe that cer- 
tain passages in the Ch^ang-an fu were written with the prospect of their 
meeting her dour, prudish eye. The two poems would seem to have been 
written mainly, but not completely, in his years of seclusion, after he had 
stretched the wings of his mind with his scientific studies. Further, his 
Loyang fu reveals an intimate acquaintance with the downward trend in 
court morale, and that points to his having had some years at court. 

[ 22 ] 



BIOGRAPHICAL DETAILS 

and was out o£ a post for five years. The grossly superstitious 
tendencies in the Wei Shu ("Woof Books") ^° aroused his 
special scorn, and he expressed this bluntly in a memorial 
to Emperor Shun. Yet he became a gentleman-in-attendance 
on the Emperor and even a confidential secretary. The eu- 
nuchs were afraid of what accusations he might make against 
them, and they made opportunities to slander him. This 
shook his nerve, and he found relief in composing his great 
Ssu HsUan Fu^^ The Emperor's confidence apparently was 
not withdrawn. 

In 136 Chang Heng was appointed Minister in the prince- 
dom of Ho Chien where the prince's disgraceful lawlessness 
had encouraged the lawless behaviour of a number of rich 
and powerful families. On his arrival Chang Heng acted 
with the utmost promptitude and courage, and the leading 
offenders were imprisoned. The princedom was brought to 
order, but Chang Heng felt the burden of his years and 
petitioned for leave to retire. He was brought back to Loyang 
and appointed a secretary in the Secretariat. Shortly after 
he died aged sixty-two (western reckoning sixty-one). 

The Biography records "thirty-two pieces" of Chang Heng's 
composition: poems, ju^ inscriptions — a very miscellaneous list. 

^^ Apocryphal supplements to the Scriptures, mostly written during the 
Han dynasty. 

^^ The internal evidence of the Ssu Hsilan Fu raises doubt as to the strict 
accuracy of this dating- as given in the Biography. Rather, the poem would 
seem to represent the frame of mind which brought Chang to resign and 
go into retirement. That was earlier in Emperor Shun's reign. The follow- 
ing features of the poem are surely significant: (i) The poem is written 
on the pattern of Ch'ii Yiian's (third century B.C.) Li Sao. (2) Chang 
Heng, like Ch'ii Yuan, takes an imaginary journey about the earth and 
in the sky. (3) Whereas Ch'ii Yiian at the "Gate of Heaven" is barred 
entry, Chang Heng is welcomed and comforted. (4.) Whereas Ch'ii Yiian's 
poem ends in despair, Chang Heng ends by finding peace and inspiration 
in his studies. 

An alternative possibility seems to be that late in Emperor An's reign he 
was besieged with doubt as to the right course to take and finally made 
up his mind to retire. This would account for the Biography's mysterious 
detail of his appointment lapsing. At the same time it would go flatly 
counter to the Biography's assignment of date. 

[ 23 ] 



BIOGRAPHICAL DETAILS 

A considerable part of these have survived, among them 
scientific works to be found in Yen K'o-chiin's Ch^ilan Han 
Wen.^^ There was also a commentary on the Chou La which 
has not survived. Chang Heng had plans for a work on Con- 
fucius' theories on the "Changes" Scripture (7 Ching)^ and 
sought permission to complete the unfinished work of the 
court historiographers on the previous reigns. None of this 
planning came to anything. Amongst his works the most 
treasured has been the ^'Fu on the Two Capitals" preserved 
in almost perfect textual condition in the Wen Hsilan — as 
also has the Ssu Hsuan Fu. His importance as a writer on 
scientific subjects has also been known to a few specialists in 
these matters. It is only in the last thirty years that his out- 
standing significance has been recognized as one, perhaps the 
chief, of China's first real scientists. 

For the four fu on the Two Capitals which are the heart 
and soul of this study, the text used has been that in the 
Ssu Pu Ts^ung K^an edition of the Wen Hsilan, For textual 
variations I have relied on the critical apparatus prepared by 
Ku Kuang-ch4 (i 776-1 835) and attached to the Hu K*e-chia 
edition (1809) of the Wen Hsuan, 

■••^ The Chilian Han Wen includes 9 /«, some over 1000 words long^. For 
the scientific works, on which I am unable to give an opinion, the reader 
is referred to Science and Civilisation in China by Dr. J. Needham and 
Dr. Wang Ling. 



[ 24 ] 



Chapter III 



PAN KU'S ''FU ON THE 
WESTERN CAPITAL" 

<><x><>c><><><><^e«><><i><>0<^ 

Preface 

One opinion is that the fu has come down from the poetry 
of antiquity. 

In the old days [of Chou] . . . when the royal unction 
was exhausted, shih [songs] were not composed.* 

In Han times at first no one had leisure for good writing, 
but in Emperor Wu's time and then Emperor Hslian's there 
came a new respect for ritual and music and for the achieve- 
ment of literary excellence. In this the common people took 
delight. Famous songs (four specified) were sung at the im- 
perial sacrifices J reign years were named after poetic objects 
of happy augury. 

The result was that those servants of the Emperor of 
the kind whose powers of language were of assistance to 
him, Ssu-ma Hsiang-ju and . . . [six others specified], 

morning and evening debated and took thought, daily 
and monthly made offerings [of compositions] j 

whilst great officers of state. Imperial Archivist Ni K^uan 
and . . . [four others specified] from time to time 
composed. 

There were those who made themselves channels of 
popular feeling, and so communicated satirical remarks, 
others who published abroad the emperor's virtus and 
thus perfected their loyalty and filial piety. 

"^ Note to the Reader: In Chapters III-VI the indented material is direct 
translation from the four fu\ the remainder, set full width, is summary and 
paraphrase. 

[ 25 ] 



PAN KU 

Here was something of like significance for posterity as the 
songs in the "Odes." Consequently in Emperor Ch^eng's 
reign these modern compositions "were assessed and entered 
up, and of those presented to the throne there were over a 
thousand pieces." The brilliance of Great Han's literature 
thus became evident, "equal in repute to that of Hsia, Yin 
and Chou." 

To go further, although both the Way and scholarship 
have their vicissitudes, yet the principles by which Virtus 
is established do not vary. 

Explore high antiquity and it was like that, examine the 
Han House and it was like this. 

Unimportant though this matter (of ju writing) may be, yet 
the new Han as well as the old literary models ought not 
to be lost. 

Your servant has observed that with peace on all sides and 
extensive buildings and embellishments in and around the 
capital, Loyang, there are still old men in the West who 
resent these developments. They boast of the ancient insti- 
tutions of Ch^ang-an, aiming to disparage the city of Loyang. 
Consequently I have composed these fu poems on the Two 
Capitals in order to expose the errors in these men's minds 
and to confound them by the beautiful ordinances of the 
present. The wording of the fu is as follows: 

I. Introduction of the Dramatis Personae 

There was a guest from the Western Capital who asked 
his Eastern Capital host saying: "I have heard that at 
the beginning of August Han when plans were made, 
the purpose was to make Loyang the capital: 

that then they stopped and went no further, going west 
and building our Supreme Capital there. 

Have you, my host, heard of the ground for this, and 
have you ever surveyed the state and order of Ch^ang- 
an?" 

[ 26 ] 



FU ON THE WESTERN CAPITAL 

The host said: "I have not. I pray you, set forth your 
store of cherished recollections, disclose your hidden feel- 
ings with their thoughts of the past: 
enrich me with [knowledge of] the August Way, broaden 
me with [the glories of] the [former] capital of Han." 

2. The Tofografhy and Geografhy 
of the Site 

The guest then describes first the topographical position 
of Ch*ang-an and its adjacent territory with mountains guard- 
ing it to the east and west, girdled by three rivers, the Yellow 
River, the Ching, and the Wei. Thus the site was fertile 
beyond all other sites in that northwestern region, and at the 
same time it was safe from outside assault: "Heaven and 
Earth's outstanding site." The Chou Sage-Kings elected to 
rule from here and so emerged like the dragon j and here 
the Ch'in (tyrant) gloated like a tiger (over its prey). Then, 
following the founding of the Han dynasty, 

Heaven and man being here in responsive harmony, 
royal intelligence was displayed: . . . 

. . . favor was extended to the West, here in very truth 
the capital was built. 

There followed the continuous prosperity of twelve successive 
reigns. Along with that went lavish government expenditure. 

3. The City and Its Suburbs 

The walls were [as] of iron, a myriad spans in extent; 
the encircling moat dug deep as an abyss. 

Three stretches of highway were laid out, twelve gates 
of ingress and egress were erected. 

Within were the streets and cross-streets, the ward gates 
numbered a thousand. 

Nine market places were opened, the merchandise dis- 
played, kind by kind in ordered rows. 

[ 27 ] 



PAN KU 

[The throng was so great] men could not face about, 
the carriages [so many] they could not make a turn. 

The town was so full that it overflowed into the suburbs, 
with a hundred stores on either side: . . . 

The red dust rising everywhere, the smoke [of the fires] 
linking in clouds. 

Thus numerous and thus rich was the populace, its gaiety 
without limit. . . . 

Its knights for hire looked like dukes and marquises, its 
shopgirls more sumptuous than great ladies. 

In the spacious environs of the city were the quarters for 
the aristocracy, and there silken tassels and (princes') crowns 
were to be seen, and carriage canopies were as thick as clouds 
in the sky. More especially was this so near the imperial 
tombs where high provincial dignitaries and the merchants 
enriched by the trade of five provincial capitals had their 
residences. This was a device of imperial policy: "to strengthen 
the trunk and weaken the branches." 

4. The Imferial Demesne 

Beyond the city's environs lay the imperial demesne of a 
thousand U with its soaring hills, its forests and shady val- 
leys to the south, with Lan T4en and its beautiful jade. 
Water was abundant, so there were fruit orchards and plant- 
ings of sweet-smelling herbs and fragrant flowers. To the 
north were the Nine Peaks, partnered by the Dale of Kan 
Chilian ("Sweet Water Spring") with its spirit-inspired palace 
buildings {ling kung). There also were the irrigation canals 
(engineered) by Cheng and Po, sources of clothing and food 
(for the people), and fifty thousand marked-out plots of 
arable land in which were grown the five cereals and mul- 
berry trees and hemp in abundance. To the east were larger 
canals carrying boats, linking up with the Wei and Yellow 
rivers, and so threading the mountain ranges and lakes and 

[ 28 ] 



FU ON THE WESTERN CAPITAL 

finally mingling their waters with the waves of the sea. To 
the west was the imperial hunting park together with en- 
closed pleasances, with fishponds and stretches of unculti- 
vated low-lying lands, containing in all thirty-six buildings 
of one sort and another for the entertainment of the court, 
the whole area being more than four hundred // in circum- 
ference. Here were to be seen unicorns from Chiu Chen 
(Annam), horses (of a famous breed) from Ferghana, rhi- 
noceroses from Huang Chih (three thousand li away, say 
the commentators), and birds from T4ao Chihj and strange 
species from outlandish countries thirty thousand // away be- 
yond the K'un Lun Mountains (Northeri: Himalayas) and 
the great ocean. 

5. The Emferor^s Personal Dwelling-Places 

The buildings of the imperial palace in their formation 
were made symbols of the heavens and the earth, their 
cross-lines in keeping with the Yin and the Yang. 

They were situated to conform exactly to spirit [emana- 
tions] from the earth, patterned after the round and the 
square of the T^ai Wei and Tzu Wei [constellations]. 

Here were ornate gate-towers reaching up into the sky, 
here crowning the hill were the vermilion-coloured halls. 

Built of rare materials, its rafters were shaped like drag- 
ons' pinions, its pillars had bases constructed of jade, 
the rich interlacing of colors making an impression of 
luminous splendor. 

The palace being on a hill, the approach was by three flights 
of steps with balustrading at the top. Within, the rooms com- 
municated, having either large or small doors to give access. 
There were sculptured pillars in the central court and rows 
of metal statues of men at both sides of the main entrance. 
Other separate palaces, sleeping apartments, belvederes, etc. 
were studded here and there on and below the precipitous 

[ 29 ] 



PAN KU 

summit: the Cool Pavilion, the Hall of Proclamation, the 
Warmth (in winter) Apartments, the Fairy Pavilion. . . . 
It is impossible to discuss them in detail . . . everywhere 
places for resting or feasting. 

6. The Rear Palaces 

The Rear Palaces (i.e. the women's apartments) consisted 
first of two buildings, one for the empress, and one for the 
first-rank concubines, and then Joy-in-Concord, Crowned Walls 
. . . (twelve buildings in all mentioned by name) j but Blazing 
Sun built by Emperor Ch^eng was especially fine. Its cham- 
bers did not reveal the building materials, the walls did not 
disclose their shape, the surfaces being covered with embroi- 
dered silk hangings. Pearls, gold, jade discs, kingfisher 
feathers, and precious stones adorned the walls. The hall 
itself gleamed with precious stones, with coral and precious 
woods. And there were the ladies, with red silk sleeves flut- 
tering in the breeze, with their flowered silks of various 
hues . . . they looked like goddesses. There were fourteen 
degrees of rank housed in these buildings and the numbers 
of inmates had to be counted by hundreds. 

7. The Administration Offices, the Archives, 
and Other Buildings 

In the buildings to left and right were the locations for 
the hundred officers of state: 

where Hsiao [Ho], Ts^ao [Ts'an], Wei [Hsiang], Ping 
[Chi], [famous statesmen] made their polices for the 
crown. 

As they lent strength to the [heaven-given] mandate, so 
they handed down the sovereign rule, and as they gave 
wings to the government, so they made civiHzation [for 
the people], spreading abroad the magnanimity of Great 
Han, removing the poisoned stings of derelict Ch4n. 

This caused the people here [in the city] to sound forth 

[ 30 ] 



FU ON THE WESTERN CAPITAL 

the voice of musical accord, to compose the song "Write 
the Word for One." 

Their merit and their virtus shone in the founding em- 
perors, the unguent of their favor imbued the masses. 

Here also were the T4en Lu and Shih Ch^u [buildings], 
where the archives were kept, where venerable elders 
patient in instruction, famous Confucians and learned 
scholars were commanded to expound and discuss in re- 
lation to the Scriptures [lit. Six Arts], to scrutinize and 
collate in relation to the disagreements and agreements. 

Here also were the Ch^eng Ming and Chin Ma Chambers 
where the work of composition went on, where fine poets 
and men of penetration congregated. 

Having gone to the roots of things, and having exhausted 
[their powers of] seeing and satiated [their powers of] 
hearing, they unfolded sectioned compositions, compared 
and regulated abstruse patterns of diction. 

Beyond all these buildings there were on all sides other build- 
ings housing functionaries of the court. 

Here were gathered the officers of rites and ceremonies, 
men who had come first in the examinations, the upright 
and the filial from a hundred prefectures: [also] officers 
of the guards, keepers of the robes, controllers of eu- 
nuchs, guardians of the gates, halberdiers of the throne 
room — a hundred different offices, each with its pre- 
scribed duties. 

There were the thousand rows of huts for the garrison of 
the Palace, whose patrols crossed in all directions j there was 
the well-planned route of the imperial chariot j and there 
were the flying galleries which led from the Wei Yang Pal- 
ace via the Ming Kuang Hall (where memorials were de- 
livered) to the Chiang Lo Palace, and from thence crossed 
the western city wall to the Chien Chang Palace outside. 

[ 31 ] 



PAN KU 

8. The Chien Chang Palace and 
Its Affurtenances 

The description runs along without very discernible para- 
graphs although changes in rhyme and the length of the 
sentences mark the movement of the narrative. 

This palace had been built with close attention both to 
artistic workmanship and the creation of auspicious influences. 
It had a thousand doors and its ten thousand windows had 
shutters which could be opened and shut. In the soaring 
interior of its main hall the roof beams went higher and 
higher so that it overtopped the Wei Yang Palace, and both 
it and the other three halls and the T'ien Liang Palace were 
cunningly devised so as to catch the light of the sun. The 
adjacent high terrace was such, reaching up into the clouds, 
that even though one were light and agile, it would be as 
if, with eyes startled out of focus, one could not ascend. So 
also with the Ching Kan Tower: 

... in ascending it before one is halfway in the ascent 
of the tower, the eyes swing round and the attention is 
astray. 

One loses hold of the railing and thus misses its support, 
is about to fall headlong but yet held back. 

In bewilderment of soul from loss [of sense] of distance 
— round one turns in one's tracks and goes down step by 
step to the bottom. 

If, in spite of fears, the top of the tower is reached, the 
descent by the dark spiral staircase is equally daunting. 

As for the two lakes in front and behind, they had three 
islands [the names having fairy connotations]. 

Here were undying plants which bloomed in winter, 
spirit trees in verdant groves: jagged rocks and beetling 
crags, an imposing spectacle of metal and stone. 

Here [statues of] Immortals placed high overhead on 

[ 32 ] 



FU ON THE WESTERN CAPITAL 

metal columns stood holding bowls for dew: a reaching 
out beyond the defilement of [this world's] clogging dust 
to gain the limpid elixir of pure, transparent ch^i [con- 
stituent gas or essence of the physico-spiritual universe]. 

He [the emperor] was carried away by Wen Ch^eng's 
windy verbiage, transported by the principles that Wu Li 
laid down: with the intent that he might consort with 
(Ch%) Sung [Tzu] and [WangTzu] Ch^ao [magician- 
immortals of legend] and their like, and they from time 
to time keep company with him in this very court. 

An abode for immortals — not a place where we mortals 
can find peace. 

9. Th^ Imperial Hunt 

And then there were the imposing spectacles of the re- 
views, the martial displays in the hunting park, useful for 
the intimidation of barbarian tribes. When the imperial order 
went out, markers were set up and the beaters drove bird 
and beast in those directions, whilst troops of soldiers encir- 
cled those areas. The Emperor and his attendant officers 
having been driven to prepared lodges, there six bands of 
hunters started the final drive. They terrified the game with 
thunderous hangings and lightning-like flashes j hills and 
pools were obliterated in the process. Then the slaughter 
began with arrows and spears and swords wielded by the 
soldiers and their officers, a wind of feathers and a rain of 
blood covering the ground and darkening the slcy. Skilled 
hunters did prodigies of daring, killing maddened tigers and 
rhinoceroses with their own hands. Lions and bears and ele- 
phants were among the bag. Nothing was left of the grass 
and the trees, the birds and beasts were all massacred. 

10. After the Hunt 

Then the Emperor went up to the Shu Yu lodge, from 
there to survey the heaped-up masses of the dead, the trophies 
of his soldiers' valor. Having enquired as to who had spe- 

[ 33 ] 



PAN KU 

daily distinguished themselves, he awarded prizes to them. 
The provision and wine carts then arrived and a quaffing 
from goblets was followed by a barbecue of the freshly-killed 
game, this by the light of torches. After the feast everyone 
relaxed, some by the shady trees and fragrant plants at the 
K^un Ming pool. Rare birds were to be seen there (eight 
names of species given), like clouds collecting, like mist 
dispersing. 

Meanwhile the palace ladies had arrived in their carriages 
and begun to disport themselves in decorated dragon- 
shaped barges floating like the wind and dispersed among the 
hundreds of adjacent buildings. With provisions there in 
plenty the festival air continued. Finally offerings were made 
to the spirits above and below, and the company returned 
singing various kinds of songs. 

II. The Western Cafital Guest Sums Uf 

Throughout this period the provincial capitals looked 
across to each other, the townships were linked one to the 
other: satrapies relied on the foundations of tens of gen- 
erations, families carried on the inheritance of a hundred 
years: scholars gained their living from the repute of 
their ancestors' [lit. ancient] virtues, peasants served the 
irrigated acres of long-tilled fields : merchants sold what 
their families for generations before had sold, craftsmen 
used the compasses and measuring square of their fathers 
before them. 

Everything fresh and abundant: everyone in his right 
place! 

An insignificant person like myself, looking for remains 
on an old scrapheap, and listening to the tales of old 
men does not get one part in ten, indeed cannot present 
a comprehensive [picture]. 



[ 34 ] 



Chapter IV 



CHANG HENG'S "FC7 ON THE 
WESTERN CAPITAL" 

I. The Introduction of Two Interlocutors 

There was a nobleman Stayed-on-Vanity j he had a super- 
cilious mind and an arrogant carriage. 

In a gentlemanly way he liked to be learned about an- 
tiquity, [and] was versed in the old historiographers. 

Hence, having much information about former genera- 
tions, he had a conversation with a teacher by the name 
of Where-did-he-Live [i.e. Anonymous]. 

He said: If a man be living in a yang time [of good 
fortune], then he is comfortable: if he be living in a yin 
age [of ill fortune], then he is miserable. 

This is a matter connected with the class [of facts] called 
Heaven. 

If a man's dwelling-place is on fertile soil, then he is in 
easy circumstances: if it is on lean land, then he is bur- 
dened with toil. 

This is a matter connected with the class [of facts] called 
Earth. 

To be miserable is to be short of pleasure, to be burdened 
with toil is to be straitened in [the] kindness [one can 
give or receive]. 

Rarely are there men who are able to go counter to these 
[conditions]. 

In small matters this is certainly so: in large matters it 
is also bound to be so. 

[ 35 ] 



CHANG HENG 

Hence emperors accord with Heaven and Earth with a 
view to perfecting culture, the masses accept instruction 
from above and thereby fashion custom. 

Since [the two] bases of culture and custom have an ad- 
justing effect on each other, how do we verify these [mat- 
ters] ? 

The Ch^in line having their base in Yung [a rich soil 
area] became strong, Chou being in Yu [a poor soiled 
area] became weak. The first emperor of Earlier Han 
made his capital in the west, and was [? could be] lavish j 
the first emperor of Later Han dwelt in the east and was 
[? had to be] frugal. 

As to the rise and fall of governments, it is invariably 
according to these factors. 

You, Sir, may I say, have not looked into Western Capi- 
tal matters: allow me to present them to you. 

2. The Tofografhy and Geografhy of Ch^ang-an 

The cultured nobleman first gives a description of the re- 
gion. There is the moral, that the site of the city is a natu- 
rally protected area and the valleys round are fertile, some 
with fields of the highest grade {shang shang). The area con- 
stitutes a "hub" of auspicious influences. According to an 
ancient legend, "Ta Ti" (Supreme Sovereign) invited the 
ruler of the Ch'in state and having feasted him, and being 
himself well liquored, very kindly made him a gift of a gold 
tablet as deed of ownership over this fertile area. This ac- 
counts for the amazing fact that Western Ch'in was able to 
unite all the states (the seven powerful states of Warring 
States China). Then when Emperor Kao Tsu entered this 
area the stars gave evidence of this being the right place for 
the capital,^ and when Lou Ching, taught by Heaven, urged 

^ Chang- Heng's actual wording is "five luei [planetary orbits] converged 
so as to lodge in the [constellation] Tung Ching." This precision of lan- 
guage equates this passage to that in the "Acts of the Emperors" under Kao 

[ 36 ] 



FU ON THE WESTERN CAPITAL 

him to go west, did he not repress his longing for his native 
place in the east. 

3. The Palace Buildings 

First the city walls were built. Then, inside the Wei Yang 
Palace was elaborated the Tzu Wei Palace with its long 
ridge-beams all colors of the rainbow and carved pillars on 
jade pediments. A gently graded road led up to it. "Double 
doors were reinforced to guard against the treacherous." It 
was planned to face four-square to the four points of the com- 
pass whilst the other buildings clustering around it (eight 
are designated) were like the stars encircling the North Star. 

Within were the ushers in constant attendance, to take 
orders and do service to the emperor. 

Without were the Lan T^ai and Chin Ma [buildings] 
where [officials] were by rotation on duty for the night. 

Next came the T'ien Lu and Shih Ch'u [buildings], the 
places where the collation of texts proceeded. 

4. The Rear Apartments 

Among the Rear Palaces were . . . (eight designated) in 
which were "bevies of modest young women, beautifully be- 
decked — how one would admire what one saw if one looked." 
The furnishings there were profuse, richly carved and orna- 
mented, "kingfisher feathers and pearls strung together with 
fine jades . . . and pearls of different hues like candles in 
the dark." What with flying galleries communicating and the 
art of skilled craftsmen carried to perfection, here was the 
place for unceasing delight. Wherever the Emperor went, 
there the music stands were ready and a feast prepared. "If 

Tsu's first year. The student should note Dubs' investigation of the alleged 
convergence (Dubs, History of the Former Han Dynasty^ I, 1 51-15 3). 
Apparently the Han records (see "Record of History," 37) were in error 
astronomically to the extent of one to two years. As far as Chang Heng 
is concerned in this matter, what he stated was that Kao Tsu assumed im- 
perial authority when he entered the Kuan Chung area and won a victory. 
That would seem to be true, whatever the actual year of the planetary 
convergence. 

[ 37 ] 



CHANG HENG 

he stayed to the end of his days, forgetting to return to his 
chamber, [the pleasure] would be inexhaustible." 

5. The Outbreak of Megalomania 

Yet such was [the sense of] the divine beauty appertain- 
ing to great monarchs, the fear was lest the difference 
between high and low should not be marked. 

Spacious as the palace area was, their hearts' desires knew 
no bounds. They yearned to outdo the Ch4n Emperor, so 
the old lodge in the Kan Chilian (Sweet Spring) healthy 
up-land was rebuilt as a summer residence . . . (three palaces 
specified). And on the range above was built a tower, Sky- 
Communicating Tower, calculated to excite awe: so high that 
birds could not reach the top of it, so thrilling, for as one 
leant on the balustrade, one heard the thunder reverberating 
below. 

6. The Chien Chang Palace Areay with Its 
Sfirit-Revealing Tower 

The Po Liang Terrace (built by Emperor Wu in 1 16 b.c.) 
having been destroyed by fire (in 104 b.c), a barbarian 
shaman advised on magic techniques. So the Chien Chang 
buildings were erected twice as gorgeous as those of the Wei 
Yang. The craftsmanship was of a high degree of virtuosity 
{kuei wet) J with its achievement of revolving weathervanes 
"hovering to the breeze." 

The Shen Ming [Spirit-Revealing] Tower reared itself 
aloft and the Ching Kan Tower, in a hundred piled-up 
storeys, 

One storey succeeds another, as you go up and up — you 
catch sight of the North Star, and are thrilled with de- 
light. 

Rising free from the world's dust into the upper air. 

You spy out the long back of the curving rainbow: you 
study the leaning together of the Hyades. If you go on 

[ 38 ] 



FU ON THE WESTERN CAPITAL 

to the flying balcony and look beyond, you observe 
straight before you Jade-Light and String-of-Jade. 

You are about to go on — ^but before you have gone half- 
way you are shaking with fear, filled with apprehension. 

Without the Tu Lu's [climber's] agility, who can climb 
high and keep going higher? 

7. The Chien Chang Area (continued) 

Four lofty and spacious halls are specified with rows and 
rows of roof beams and the eaves elevated so that light could 
penetrate. Then there was the Sky-Bridging Palace with a 
portico so huge that with only one leaf of its door open a 
four-horse carriage with banners flying could drive through 
with a loose rein. It was such a medley of corridors and 
pavilions . . . "with a hundred doors and a thousand win- 
dows," with such a medley of interlacing paths that "as you 
gaze, you cannot find your way back." And then there was 
the Chen T'ai Terrace dipping up and down as it went east 
to cross the city wall, making a private means of communi- 
cation, with the watchmen always on guard. 

Two spacious lakes were there with islands on which were 
replicas of the fairy mountains, Ying Chou, Fang Chang, and 
P^eng Lai, arranged in a pattern with the Hunting Park 
rising to view beyond. And round about were herbs of mystic 
potency, and in the lakes hugh fish which sometimes got 
caught in the shallows. 

After this fashion he [Emperor Wu] selected Shao 
Chun's doctrines and believed in Liian Ta's dogmatizing. 

He held that with such techniques a man's natural life could 
be extended. 

He admired [the immortals] of the past, [Ch'ih] Sung 
and [Wang] Ch^iao, met Hsien Men in the roads of the 
sky: he thought of mounting the dragon from the Lake 

[ 39 ] 



CHANG HENG 

of Tripods [as the Yellow Emperor was said to have 
done] — how could he care for conventional ambitions? 
But if it was really possible to live from generation to 
generation, why such feverish activity in building his 
mausoleum? 

8. The City and Its Suburbs with 
Their Inhabitants 

We pass now to view the city with its walls and three 
gates each admitting twelve carriages and its criss-crossing 
streets, with its building plots evenly spaced and its raftered 
dwellings in orderly rows. In the northern quarter were the 
highest grade houses, stoutly built by the very best crafts- 
men, "with the timbers adorned like the silk of robes, with 
the earth [walls] colored in red and purple," and stands of 
weapons from the imperial armoury. "Unless he were a Shih 
[Hsien] or a Tung [Hsien] [notorious palace favorites], 
who could dwell here?" 

For the nine markets there were official guards. "Those 
who sold goods made a double profit, those who sought goods 
did not [go away] empty-handed." Actually merchants and 
pedlars of all degrees, (especially) the men and women 
hawkers, sold inferior goods mixed with those of high qual- 
ity, "deceiving stupid rustics, and making fools of small-town 
folk." 

The merchant families were "dressed as luxuriously as 
the scions of the imperial house." As for the families of Weng 
Po, Cho, Chih, and Chang Li (well-known millionaires of 
the time), "bells were struck [for meals] and they ate out 
of ting [cauldrons]. . . . The dukes of the Eastern capital 
[today] cannot better them in appearance." The "wandering 
braves," men who were reckless of death, made gangs among 
themselves, and had followers "like clouds . . . quick to quar- 
rel, roar — of a sudden, like tigers, like wild cats." As for 
the scholars from the country round, ready to debate any 

[ 40 ] 



FU ON THE WESTERN CAPITAL 

topic, they were in every street and alleyway, "splitting hair^, 
tearing muscles apart. . . . Those whom they liked grew the 
fur and feathers [of a career], those whom they hated, grew 
the boils and bruises [of no career]." 

9. The Hunting Park 

The Imperial Demesne \_chi'] of a thousand U was under 
the control of the Intendant of the Capital. 

Within the area were one hundred and forty-five pleasure 
buildings of various kinds. North, south, east, and west were 
. . . landmarks. Within this area was a preserve for birds 
and beasts and fish. Of trees there were . . . (eight named) 
growing thickly. ... Of herbs there were . . . (thirteen 
named), with dwarf bamboos growing luxuriantly in clumps. 
Endless were the hills and valleys and level pastures. There, 
also, was the dark pool of K'un Ming, encircled by an em- 
bankment which was planted with willows. There were all 
sorts of strange species of fish in it. With regard to the birds 
(five species named), "at the beginning of spring they were 
ready to arrive, at the end of autumn they sought a warm 
climate: so many different shapes and voices that it is im- 
possible to describe them." In the winter the foresters pre- 
pared the terrain for the hunt. They burnt over the woods, 
cleared away the timber, and cut back the brambles. 

10. The Imperial Hunt 

The Son of Heaven came riding in his six-horse carriage 
with the appropriate symbolic appurtenances. Accompanying 
him were a thousand carriages thundering along and ten 
thousand horsemen galloping like dragons. 

It was not only play, for there were books of magic and 
lively tales, based on Yii Ch'u's nine hundred [chapters] : 
what was wanted in relaxation, all waiting ready. 

Thus it was that Ch^ih Yu grasped his halberd, spread 

I 41 ] 



CHANG HENG 

his shaggy coverlet [a tiger skin], a guard against aught 
uncanny. 

By this means was known the malignant [tricks] of the 
spirits among the mountain and water demons, not one 
could cross the emperor's path. 

The imperial guards were on parade at two points, the 
torches lighted, the drums a-thunder. The hunt was on. Birds 
and beasts were completely panic-stricken and in their terror 
dashed themselves against the carriage wheels, having no- 
where to take cover . . . 

No arrow was fruitlessly discharged, no dart flew in vain. 

The stiff corpses of the slain, bird and beast, glistened 
like tide- washed pebbles [on a beach]. 

And there to view were the strung-together [victims] of 
the traps, battered to death with staves and poles . . . 

Before the sun was down and [the shadows] come, the 
killing was done, out of ten, seven, or eight. 

If the pheasants or crafty hares were so swift as to escape, 
there remained the swift of wing and the light of foot: they 
were bound to be struck down into the falconer's glove or 
be crunched by a dog at the end of a leash. 

As for the savage indomitable beasts, with bristles erect 
and eyes glaring out of the corners of deep sockets, the terri- 
fying ssii (rhinoceros) and tigers whom no one could with- 
stand — there at work were the professional huntsmen of the 
kidney of Chung Huang and the stamp of [Hsia] Yu and 
[Wu] Huo, with red fillets in their bristling hair, bare-armed 
and spear-handed, advancing and circling, striking and grab- 
bing. They could plunge through bramble-brakes and tear 
fences apart. High-hearted and nimble, they could dive into 
caves and snatch a fox from its hole, dash up overhanging 
hills to catch the wild ass, or swing out to the end of a branch 
and seize hold of a monkey. 

[ 42 ] 



FU ON THE WESTERN CAPITAL 

II. After the Hunt 

At this time would come the ladies of the harem, laughing 
and chattering, delighted to be there. After gazing their fill 
at the dead birds and beasts — and turning round for another 
look — they would repair to the Ch'ang Yang lodge. 

Those who hunted on foot were rested and the horses 
and chariots were paraded. 

The corpses of the birds and beasts were put in piles, 
their numbers counted, many or few: the plunder was 
apportioned out, victor's bounty carefully awarded. 

The regiments were drawn up in a thousand rows, and wine 
and meats were distributed in the light of torches. A steward 
came on horseback to inspect. Such was the Emperor's bounty 
"the runners and drivers were pleased, the soldiers forgot 
their fatigue." 

The carriage-officer ordered the horses to be reharnessed 
and the carriages to turn and move over to the right. 

They wandered by the Wu Tsa lodge, they circled or 
stood by the K'un Ming lake. 

The Emperor and his entourage would climb the terrace on 
the island there and proceed to shoot at high-flying wild geese 
and cranes. Then they would go to the Warden of Boats to 
provide entertainment. Barges with feathered sides and covers 
put out into the lake : "the oarswomen breaking into song, one 
leading, the others coming in with a chorus." Also the Huai 
Nan and Yang O action-songs were performed, and water 
demons and snakes frightened away. Then fishing started 
with hooks and nets, including drag-nets — but since the War- 
den of Marshes himself broke the regulations, what did it 
matter? The spawn were exterminated and the water weed 
dragged up by the roots. 

[This] overweening lust for hunting and fishing, [these] 
exertions for capturing small fry. 

[ 43 ] 



CHANG HENG 

This ransacking and scouring, this womb-grabbing and 
egg-snatching [ch^u\ : 

Snatching [ch^il] at pleasure today, [how] have I leisure 
to pity what comes after me tomorrow? 

In a time of settled peace, how realize it [all] will totter 
to ruin? 

Then the Emperor took his ease in the midst of his priceless 
coverlets and other appurtenances of extraordinary beauty. 
Then he favored the arena with his presence. There was 
staged the "marvelous entertainment called ^chileh /i' [horn- 
butting]." Strong men balanced weights, acrobats and jug- 
glers performed (six kinds of feats noted). Choirs dressed as 
fairies sang, and men in the skins of leopards and bears 
danced, and others as white tigers struck great lutes, whilst 
green dragons played flutes. Some dancers at first would be 
like the whirlwind and then like falling snow. A colossal 
beast would appear, eight hundred feet long, and when it 
turned its back it was a fairy mountain with precipitous sides 
and bears and tigers pouncing on each other. . . . Conjurors 
would suddenly change their appearance and split their shapes, 
swallow swords and belch forth fire . . . smoke growing 
darker and darker. 

[But there was the case of] Old Master Huang [a magi- 
cian] of the Eastern Ocean who with his sword of gold 
worked spells, 

aiming at pacifying white tigers — in the end he could not 
save himself. 

He hugged his evil, poisoning magic — and thus it was 
that it did not work. 

12. The Emferor's Undignified Amour 

So then, when the final scene was staged, with masts erected 
and flags displayed, when boy acrobats flitted up and down 
between the masts, seeming to fall yet caught by the heels. 

[ 44 ] 



FU ON THE WESTERN CAPITAL 

Thus, with the completion of these many scenes, the 
mind [of the Emperor] was intoxicated. 

With the round of pleasures carried to the extreme, bore- 
dom came ravelling in his breast. 

The sentries of the gates being warned, [the Emperor J 
would creep out incognito. 

He descended to the level of the common man, and roamed 
through the suburbs. 

The dragon's ability to transform himself manifests the 
true imperial majesty. 

Thereafter he would seek a pleasure house and a compliant 
beauty, who would exercise all her languishing and provoca- 
tive blandishments. 

One ogling glance enough to topple down a city's de- 
fences! Who, whether a Chan Chi [Liu-hsia Hui, an in- 
corruptible Chou scholar] or a Buddhist monk, could 
avoid being moved? 

With fourteen ranks [of women in the harem] competing 
in fascination and grasping at the honor, favor and dis- 
favor are never constant j it is personal feeling [ai] alone 
that clinches the matter. 

[Emperor Wu's] Consort Wei rose through her luxu- 
riant locks, [Emperor Ch'eng's] Consort Flying Swallow 
came to favor by the lightness of her body. 

13. Final Conclusions 

So then — his intent full set to exhaust desire, giving 
himself up to enjoy to the full, he took warning from the 
Songs of T^ang: "If you do not enjoy your possessions 
now, another will enjoy them." 

The authority was his to create a precedent — where then 
the restraint of established procedure [//] ? 

[Emperor Ch'eng] added [to the harem] a new rank 

[ 4J ] 



CHANG HENG 

above that of chieh yiij Tung Hsien [Emperor Ai's fa- 
vorite] being already made a duke was yet further en- 
nobled. 

[Emperor Ch^eng] promised concubine Chao that there 
should be none above herj [Emperor Ai] thought to 
exalt Tung Hsien as [Sage-king] Shun was exalted by 
[Sage-king] Yao. 

[This was when] Wang Hung in attendance at the side 
protested that the Han years had been ones of peace not 
of revolutionary change. 

Emperor Kao Tsu created the heritage, his descendants 
built on his foundation. 

Labor for a season brought lasting ease, order was main- 
tained without overt action \wu wei] . 

With pleasures all there to be indulged, why should one 
worry, why should one take thought? 

For so many years was this so, two hundred seasons and 
more: merely the soil was fertile, the outlands produc- 
tive, all commodities in heaped abundance: the high 
passes bringing safety on every hand, forming a girdle 
easy to protect. 

Those who had gained them had the power j those in 
possession of them endured in the land. 

Where a stream flows far it is hard to stop, where the 
roots are deep [the tree] can hardly rot. 

The result was the dissolute temper of luxury in excess — 
a sweet odor became pungent and more and more rank. 

14. The Nobleman Makes His Bow 

Your unworthy pupil, Sir, was born three hundred years 
after [the beginning] of Han, and hearing these things 
from those who never heard them [i.e. learning them 

[ 46 ] 



FU ON THE WESTERN CAPITAL 

only from hearsay] has been uncertain as one in a dream 
and cannot grasp even one part of them. 

Today [our] Imperial Masters share the title of "Sover- 
eign August" with Heaven j they have covered the whole 
world and made it one family. 

Of all rich inheritances, there is none so great as ours. 

The pity is that [now] we are not allowed to make lavish 
elegance the country's glory. 

Yet parsimony goes with a mean spirit and forgets what 
the ^'Song of the Cricket*^ says. 

Is it not true that what we want we cannot get and what 
we can get we do not want? 

I am stupid and in error to be sure — pray let me hear 
the view by which these matters are [rightly] explicable. 



[ 47 ] 



Chapter V 



PAN KU'S ''FU ON THE 
EASTERN CAPITAL" 

C><><>0<><><><>00<><><>00^^ 

I. The Eastern Cafital Host Jibes at the Western 
Cafital Guest^s Opinions 

The Eastern Capital Host drew a deep breath and ex- 
claimed: Alas, the effect that habitat and custom have on 
people! You, Sir, are indeed a man of Ch^in, boasting 
proudly of buildings and the protection of boundaries by 
mountains and rivers. 

In truth you know about King Chao-hsiang and have an 
understanding of the First Emperor. How [then] can 
you discern the words and deeds of great Han? 

Take the First Day of Great Han : it burst forth from a 
humble commoner ascending the imperial throne, from a 
few [momentous] years initiating a myriad generations. 

This was something the Six Scriptures could not de- 
scribe, something it was impossible for the former sages 
to mention. 

At this juncture [the rising emperor] attacked the con- 
tumacious, thus conforming to [the will of] Heaven, 
punished the trouble-makers, thus following the desires 
of the people.^ 

So it came about that Lou Ching, having calculated the 
situation, offered his theory [that the capital should be 
at Ch'ang-an] j Hsiao Ho, having weighed what was re- 
quired, extended the layout of the Capital. 

^ For the problem raised by textual variants at this point see Sun Chih- 
tsu, Wen Hsilan K'ao /, ch. i, and G. Margoulies, Le "Fou^* dans le Wen- 
Siuany Paris, 1926, pp. 55-56. 

[ 48 ] 



FU ON THE EASTERN CAPITAL 

Was this because the times were flourishing and [the 
Emperor] was content with it [the decision and the pal- 
ace Hsiao Ho had built]? Indeed noj it was perforce 
the policy. 

My dear Sir, you have not discerned [this matter] 
aright J and being dazzled by the trifling accomplish- 
ments of succeeding generations, are you not in a state 
of darkness? 

2. Wang Manges Usurfation and the Salvation 
of the Country 

Now I will speak to you about the government of Chien 
Wu [Emperor Kuang Wu's opening term of rule] and 
about the events of Yung P4ng [Emperor Ming's reign], 
scrutinizing their supreme purity in order that I may 
correct your erroneous frame of mind. 

In the previous period there was Wang Mang's revolt 
and the Han line was broken midway. 

Heaven and man embarked on concerted slaughter j all 
within the six points of the universe joined in destroying 
him [Wang Mang]. In the ensuing disorder the living 
were almost wiped out, the ancestors were cut off, no 
coffin in the valley was not destroyed, no house in the 
city's precinct left intact j corpses filled the countryside, 
rivers ran with blood: 

The disasters of the war between Ch'in and Hsiang Yu 
were not half so frightful: since writing began no such 
tale has been recorded. 

Therefore the common people wailed aloud, made their 
plaint on highj the Sovereign on High [Shang Ti] in 
pity looked down, and gave mandate to the Sage Em- 
peror [Kuang Wu], 

Thus it was that the Sage Emperor made divinations and 
found the right prognostications, and then, in his indigna- 

[ 49 ] 



PAN KU 

tion, launched thunderous attacks. Having crossed the Great 
River and strode over North Peak, he named his regime and 
established his capital at Loyang. 

He restored the ravaged [works of] a hundred kings, as- 
sisted the cleansing powers of the universe. 

Fixing the beginning of a new era, he established his 
government, in very deed continuing [the act of] 
Heaven. 

He revived the tradition of [Sage-King] Yao, restored 
the line of Han, bringing food and plenty to the living, 
restoring the frontier lands of the state, his merits beyond 
all those in the past, his labors more than the labors of 
the Three and the Five.^ 

It was not that he merely unified social standards and 
paraded a pompous show of hierarchy and dealt with 
that necessity, of the present as of the past, to have one 
sage man tread the path of bringing order out of anarchy. 

He also on that First Day [of the new order], when 
Heaven and Earth were in revolution, everywhere within 
the four seas recreated husbands and wives j made a [new] 
beginning for fathers and sons. 

Sovereign and subject were for the first time established, 
the relations of man with man started. 

These were the very means by which Fu Hsi [the first 
Sage-King] founded the virtus of emperors and kings. 

So also it was with the efficacy shown by the Yellow Em- 
peror in dividing the provinces and establishing market days, 
making boats and carriages, implements and weapons: with 
the glory of royalty shown by T'ang and Wu in enacting 
Heaven's punishments. In his changing of the capital our 
Emperor emulated the middle revival of the Yin era, and 

^ There are various traditions. A contemporary one identified the "Three" 
as the Sage-kings Fu Hsi, Shen Nung, and Sui Jen, and the "Five" as the 
Yellow Emperor, Chuan Hsii, Ti K'u, Yao, and Shun. 

[ JO ] 



FU OAT THE EASTERN CAPITAL 

that of the Chou. In having at first "no inch of ground or 
man to command" he was on a par with Emperor Kao Tsu. 
In his discipline of himself and observance of tradition he 
was more reverent than Emperor Wen, in searching antiquity 
and performing the feng sacrifice on Mount T'ai he achieved 
greater glory than Emperor Wu. Thus Emperor Kuang Wu 

. . . schooled himself in the virlus according to the Six 
Scriptures, brought past ages into focus and argued their 
merits: and his humane sageful works being completed, 
the iao of the emperors and kings was perfected. 

3. The Emferor Ming^s Achievements 

Coming to the Yung P4ng (Lasting Peace) period of Em- 
peror Ming, this splendor of perfection was redoubled, show- 
ing itself in developing the lofty procedure of the Three 
Yung (the Ming T'ang, Ling T^ai and Pi Yung),^ in the 
ritual robes, in the fostering of literary refinement, in the 
display of conspicuous beauty, in the solemn gathering in the 
Kuang Wu fane. The Emperor made a royal progress 
throughout his dominions, examining everything, 

... in his own person seeing what the myriad states had 
and what they lacked, examining the extent of his influ- 
ence, diffusing the royal light that it might brighten dark 
places. 

Thereafter he enlarged the ancient city of Loyang, 

^ I have not attempted here or elsewhere to give English equivalents to 
the terms '■^ming fang^^ "ling t'aP^ and "fi yung.^^ All three were sets of 
functional buildings, and these names give no indication as to the various 
functions fulfilled. Ming t'ang^ lit, "bright hall," may signify bright with 
the brightness of the sun, or of imperial prestige, or of sage-like intelligence, 
or of widely influential moral excellence. Ling i'aiy lit. "spirit terrace," 
gives a clue to the primitive pseudo-science of astronomical observation 
dating from the time when signs in the heavens struck an animistic awe 
into the hearts of the watchers. The title does not do justice to the use 
of instruments and strict mathematical calculation in Han times. As for 
fi yung, there were alternative forms to the graphs. Pan Ku seems to favor 
the interpretation which indicates a circular moat of flowing water with 
square buildings within the circle. Another interpretation was "sovereign 
concord." 

[ 51 1 



PAN KU 

. . . fanning it out on a grand scale, in towering grandeur 
and resplendent symmetry j [thus] embellishing [his] 
Han capital in the midst of Chu Hsia [China], making 
a control for all parts of the world and forming their 
crowning point. 

Here within the royal city the houses were resplendent, 
revealing a divine artistry, which prodigality could not 
surpass, which frugality could not call prodigal. 

Outside the city he constructed a park where the rich herbage 
nourished the animals, and a lake where water plants con- 
cealed the fish. The size of it was of the modest proportions 
of the parks of antiquity. 

4. T/ie Imferial Hunting Preserve and the 
Seasonal Hunts 

In the park on suitable occasions the Emperor conducts 
the seasonal hunts, reviewing horse and foot and training 
them in martial exercises. In so doing he is sure to use the 
"Royal Regulations" and hunting songs in the "Odes" (four 
mentioned). After certain rites the carriages move off with 
the Emperor leading in his jade-adorned carriage with horses 
of the due seasonal color . . . "majestic, debonair." With the 
nature spirits in attendance the first stage of the hunt has 
begun "with a thousand chariots thundering along, and a 
myriad horsemen whirling in flights," lances sweeping the 
clouds, feathered plumes obliterating the rainbow, banners 
brushing the sky, "a brilliant scintillating display. . . ." On 
arrival at the center of the park, the three army corps are 
drawn up, each detachment taking its allotted station. The 
signal torches are raised, and the whole force of drivers, 
horsemen, and bowmen rush to meet the driven game. There 
is no time for bird or beast to escape. 

In the blink of an eye, suddenly the game carts are full, 
[for] the pleasure of the hunt is not indulged to the full, 
there is no wholesale slaughter. 

[ 52 ] 



FU ON THE EASTERN CAPITAL 

The outriders return to the roads, the carriages move off 
in due order. 

5. All Nations Turn to Loyang 

Thus it is that he [the Emperor] offers the Three Sacri- 
fices [to Heaven, Earth, and the Ancestors], presents the 
five sacrificial animals, does his duty to the gods above 
and below, cherishes the hundred dieties. 

In his audiences [with the Five Ti] in the Ming T'ang, 
in his visits to the Pi Yung, displays his glory, diffuses 
his august influence. 

Ascending the Ling T^ai, examining the favorable aus- 
pices, as he looks up and looks down to the Ch4en and the 
KW [symbols of heaven and earth] he relates the signs 
to [his] Sacred Person. 

He marks Central Hsia [China] and diffuses his virtus 
there, he scans the four frontiers and uplifts his sublimity 
there. 

West, east, north, and south goes his influence into outland- 
ish regions where the borders of one country are not neigh- 
bor to another. In lands where the might of Emperors Wu 
and Hsiian had not brought submission. 

No one on land or water but breathless and trembling 
comes running to pay homage. 

Forthwith peace comes to Ai Lao [in Yunnan] and it 
becomes the commandery of Yung Chiang. 

On the first day of the first month at the beginning of the 
year all meet in the Han capital, and 

. . . the Son of Heaven receives maps and reports from 
the four seas, accepts the tribute treasures from ten-thou- 
sand countries. 

He tranquilizes the barbarians as well as the Chu Hsia (Sini- 
cized states). So in tents in the Cloud-Dragon Court there 

[ 53 ] 



PAN KU 

are officers to assist the ch'unhou (vassal kings) to comply 
with imperial etiquette. A great feast is held, with a myriad 
vessels of wine, with goblets of gold and jade, with rare deli- 
cacies and fat bullocks. The Music Master and his orchestra 
and dancers are there and perform traditional measures. The 
barbarians also present their music from the four quarters, 
inspired by the Emperor's far-extended virtus. So with the 
Emperor in convivial mood his subjects drink well and the 
primordial vapor of harmony descends. At the sound of a 
bell the officers retire. 

6. The Joyful State of Sober Simflicity and Earnest 
Endeavour y with Schools Everywhere 

Thus it is that the Emperor, surveying the pleasure of 
joy in ten thousand places, as he bathes them in the unc- 
tion of his grace, fears lest the spirit of extravagance may 
arise and there be sloth over the staple occupation of the 
East [i.e. agriculture]. 

He revives old regulations, sends down clear fiats: pub- 
lishes statutes to exalt frugality and moderation, with 
economy to show complete simplicity. 

He discards the beautiful ornaments of the women's 
apartments, dispenses with the trappings of his carriages j 
he stops the demoralizing pursuits in which craftsmen 
and merchants [engage], he exalts the great work of till- 
ing the soil and [planting] mulberry trees. 

So then the orders go forth to all within the four seas, 
unessentials are discarded, and a return made to basic 
[needs], the meretricious is outlawed and men abide by 
the true j women weave silks and fabrics, men give them- 
selves to ploughing and weeding j utensils for daily use 
are of clay or calabash j with clothes, preference is given 
to plain white and black. 

He [the Emperor] counts the prettiness of fine silks a 
disgrace and will not wear them j he despises the rare and 

[ 54 ] 



FU ON THE EASTERN CAPITAL 

exquisite and attaches no value to them: he leaves gold 
in the mountains, pearls in the depths of the waters. 

Thus it is that the people cleanse themselves of blemishes 
and cast away impurities, and so can mirror to themselves 
the height of purity : body and soul untroubled, the senses 
inactive, the fountain head of concupiscence is extin- 
guished, and the spirit [lit. heart-mind] of modesty and 
shame is born. 

No one but is free and easy and satisfied in himself, [no 
one but] possesses the sheen of jade and the ring of gold. 

This is because within the four seas central schools are 
as a forest, country schools full to the doors: there is 
interchange of civilities, the ritual implements in good 
supply : dancing below, singing above, treading out virtus 
[te], singing over jen [man-to-man-ness]. 

At the feasts when the courtesies of bowing are ended. 
All praise the mysterious virtus m beautiful and noble 
terms. 

Their mouths are filled with concord, they breathe its 
essence forth, acclaiming the day of abundance that is 
here. 

7. The Eastern Host's Final Argument 

Nowadays people who engage in discussion only know 
how to recite the books of Yu [Shun] and Hsia, to chant 
the songs of Yin and Chou, to expound Fu Hsi's pat- 
terned lines in the Changes^ to discuss Confucius' Sfring- 
and-Autumn Annals. 

Few there are who are capable of getting the essence of 
what is transparent and what is blurred in the past and 
the present, [capable of] enquiring deeply into the way 
in which the virtus of Han came to be. 

You, Sir, are well acquainted with past institutions, but 
you are also speciously carried away by non-essentials. 

[ ss ] 



PAN KU 

To recall the past and [thereby] know the present has 
become difEcult, and those who understand the virlus are 
few indeed. 

Also, to live away there on the western frontier guarding 
the passes, how can this be comparable to living in the center 
of the realm, to which "ten thousand places converge like 
the spokes of a wheel"? The western rivers and mountains 
do not compare with those of Loyang where the Yellow 
River makes a girdle and the Lo River flows past, the (two) 
waters of the (Mystic) Diagram and the Script. The build- 
ings of Chien Chang and Kan Ch^uan, with the offerings 
there made to minor spirits (hsien), do not compare with 
the Spirit Terrace and Bright Hall where the concord be- 
tween Heaven and man is effectively controlled. . . . How 
can the Ch'ang-an pleasure ground be equal to the Pi Yung, 
encircled as it is by water which symbolizes the four seas 
and brings the riches of the Way and the virtus? The law- 
lessness of your bravos and dissipated townsfolk is not to be 
compared with the universal observance of right procedure 
(in Loyang), its rhythmic order and strict behavior (/ i chH 
cm). 

Vainly have you studied the O Pang, the Ch4n [Palace] 
that soared into the heavens j you are ignorant of how 
Loyang conforms to a plan: you know well the guarded 
pass at Hsien Ku, but you are ignorant of the fact that 
true Kingship has no frontiers. 

8. The Guest^s Discomfiture 

The host's speech was not concluded when the Western 
capital guest, his eyes jumping from right to left, being 
out of countenance, shrank back and round to descend 
the steps in plain fear; minded to take his leave, he held 
out his hands in farewell. 

The host said, "Go back to your place, I am going to 
give you five pieces of verse [which I have composed]." 

[ 56 ] 



FU ON THE EASTERN CAPITAL 

The guest, having finished reading, exclaimed, "How 
admirable! 

"These verses are in their main ideas more right than 
Yang Hsiung, in relation to their matter more real than 
Ssu-ma Hsiang-ju. 

"Not only are you, my host, most erudite, but also you 
have lighted on this happy age. 

"Child that I am, I talk wildly and disrespectfully, I do 
not know how to discriminate. 

"Now that I have heard the True Way, I beg leave to 
chant these verses all my days." 

9. The Five Odes 
The Odes are as follows: 



Hail resplendent Ming T^ang! Ming T'ang of [heav- 
enly] radiance! 

The Sage Sovereign [Emperor Ming] did sacrifice, rev- 
erently, majestically! 

[Here] the High Gods are feasted, at five stations in 
timely order. 

Who is f'ei [ancillary] to them? Late-Ancestor Kuang 
Wu. 

[Under] universal Heaven from end to end of Earth, 
each [dignitary] assisting according to his ofiice! 

How superb, how continuous, verily bringing many bless- 
ing! 

PI YUNG 

Here flows the Pi Yung with its rippling waters. 

The Sage Sovereign arrives, a bridge of boats made there 
[for him to cross]. 

[ 57 ] 



PAN KU 

The silver-haired veterans of the state, [whom he treats 
as] fathers and elder brothers! 

What strictness of deportment, what brilliance of filial 
piety and friendliness! 

How majestically the All Highest reveals the ways of 
our Han! 

Their transforming power so wide, so god-like! To all 
time may we see their success! 

LING T^AI 

And here one traces the Ling T'ai, the Ling T'ai long 
venerated! 

Careful to time the Emperor ascends, here he examines 
the proofs of blessing. 

Sun, moon and stars disclose their essence, the five forces 
display their sequence. 

How gently the auspicious winds, how abundant the 
sweet rain! 

The hundred grains burgeoning forth! The plants all 
growing in profusion! 

Rich harvests again and again! What joy to the Em- 
peror! 

PRECIOUS TRIPODS 

Ah tributes from the high peaks, treasures from the great 
rivers! [May be] emitting a metallic sheen, [may be] 
beclouded with flying vapor! 

Precious Tripods appear, their colors mingled in confu- 
sion! 

Radiant with sparkling light, covered with dragon script ! 
Presented in the ancestral fane, offerings to the sacred 
spirits ! 

Giving lustre to [their] spirit virtus, for a hundred thou- 
sand years! 

[ 58 ] 



i 



FU ON THE EASTERN CAPITAL 
WHITE PHEASANTS 

Open the spirit books, unroll the auspicious drawings! 

The capture of white pheasants, the presentation of white 
crows ! 

Happy auguries abound, concentrated in the Capital. 

Ruffling their glistening wings, with tail feathers erect! 

Their appearance so clean, so pure, so exquisite! 

How they display the Sovereign's virtus, equalling 
[King] Ch'eng of Chou! 

Extending ever further in space and time, fulfilling the 
blessings of Heaven! 



[ 59 ] 



Chapter VI 



CHANG HENG'S "FC7 ON THE 
EASTERN CAPITAL" 

I. The Teacher* s Reaction 

Teacher Anonymous at this point was like one who could 
not speak: he was disconcerted for a moment. Then he re- 
gained his composure, smiled and said: 

You, Sir, are one of those called learned in non-essentials 
who have gained [only] surface impressions and have 
prized the ear and despised the eye. 

If there be the breast [only] and no mind, then there 
can be no self-restraint coming from U [the code of 
honor] . 

No wonder there is this despising of the present and 
glorifying of the past. 

Yu Yii, the poor exiled minister among the western bar- 
barians, [could yet] deride Duke Mu [of Ch4n] for 
[pride in] his buildings. 

How then, with your cherishing the past [and so] know- 
ing the new [present], discriminating the true and the 
false, could you approach these erroneous [conclusions] 
of yours? 

2. The Real Lesson of the Dynastic Revolutions 
(Third Century B.C.) 

Take the last days of the Chou dynasty. "It could not 
govern" . . . and so fell to "a tiger from the west." At that 
time the Seven Powers strove for preeminence, rivaling each 
other in extravagance. . . . Ch^in with its voracious beak 
struck from afar and won the whole area. He [the First 
Emperor] thought to monopolize luxury, counting none 

[ 60 ] 



FU ON THE EASTERN CAPITAL 

equal to himself. So he built his palaces and taxed to the 
limit, for small defaults punishing barbarously. . . . The 
poor "blackheaded ones" [the common people] went in fear 
and trembling. 

Treated as serfs 

the people unable to bear this usage, rested their shoul- 
ders against Great Han, were happy to honor Kao Tsu. 

Kao Tsu, in obedience to Heaven, issued a "great war cry" 
and swept away his opponents. He had no time, he said, to 
build another capital. His western craftsmen, "their eyes play- 
ing over O Pang," went beyond all bounds. He cut their 
plans again and again, although even so they transcended 
the Chou buildings. 

Observers despised them and called them meanj the Em- 
peror ridiculed them as too big and uncomfortable. 

3. The Crossness of the Nobleman^ s Blindness 

Moreover, Emperor Kao Tsu under heavenly authority 
established a family and remade our land of China. Emperor 
Wen, a man of personal frugality, governed by exalting the 
virtus of peace. Emperor Wu enormously expanded our ter- 
ritory and made the sacrifice on Mount T^ai. Emperor Hsiian, 
by his severity, reconciled the Jung and Ti barbarians, so 
that they came to pay homage. All the emperors kept their 
spirit tablets enshrined and served them with due seasonal 
offerings. Also, the inscribed sacrificial vessels were continu- 
ously made more and more splendid. Thus in omitting the 
Emperor's true excellencies and noting their bad qualities, 

... no wonder you have not been disgusted by the past, 
have thrown into the shade the good and displayed the 
evil J [and] in fact, my dear Sir, have been ignorant of 
[the forces of] words. 

If you must make unbridled luxury reputable, this is to 
make the Yellow Emperor's [thatched] Ho Kung and 

[ 61 ] 



CHANG HENG 

Shun's [thatched] Tsung Ch*i^ definitely not equal to 
Chieh's Yao Tower and Chou's Cheung House. 

Who, then, could have been a [Ch'eng] T'ang and 
[King] Wu, making revolution by force of arms? 

You should also make observation of the conditions in 
the Eastern Capital and thereby awaken yourself [to the 
major facts]. 

4. The Rightness of Loyang as the Captal 
for True Sons of Heaven 

Moreover, the true Son of Heaven has the Way, his 
sphere of protective power goes beyond the [four] seas. 

In protecting his throne he uses human-heartedness, he 
does not depend against injury on the defence of passes. 

If there be no good faith in the people, why speak of 
craggy mountains and a girdle of narrow passes! 

The Ch^in regime put its trust in the Two Passes, and it 
fell through their twice being penetrated. How much wiser 
to have a site central to a wide circumference! Thus an early 
king (Ch'eng of Chou) examined all the Nine Regions, and 

. . . with a measuring tablet [t^u kuei] marked the shad- 
ows, neither too short nor too long. 

There at the meeting point of the winds, and rains, he 
thereafter built a royal city. 

There with the Yellow River at the back and facing the 
Lo River . . . with the hill slopes and the valleys en- 
circling it (named details given). 

A hot-water spring issues from a red and black rock. . . . 
The Goddess Fu Fei there has her haunt, and in high an- 
tiquity first a dragon and then a turtle emerged from the 
river, bringing mystic writings. 

^ The names by which the Ming T'ang of the Yellow Emperor and that 
of Shun were known. 

[62] 



FU ON THE EASTERN CAPITAL 

Lord Shao examined the localities, only Lo was auspi- 
cious: Duke Chou began the foundations, his [? inked] 
string was straight. 

Wide roads and battlemented walls were built, a ming t^ang 
erected, and here was "a city and township in balanced pro- 
portions." The beginning of Han broke the succession. Then 
came Wang Mang and his wicked interregnum. 

Throughout thrice six successive years he took his ease 
as a thief on the heavenly throne: in that time through- 
out the nation, not one dared have an opinion other than 
his. 

The Father of our Age (Kuang Wu) in his rage at this flew 
like a dragon into battle. 

He gave battle-axes to twenty-eight generals, and the 
usurper was destroyed. 

With his (Wang Mang's) malignant influence extirpated, 
the elements were again in harmony. 

The wise man, with mystic penetration {hsUan Ian), dis- 
cerned the Capital, this Loyang. 

He said, "Abide here"j he said, "Glory will be continu- 
ous here." 

So with the Way flourishing in the realm, the ascent was 
made of Mount T^ai and the sacrifices of peace-returned cele- 
brated, and kinship claimed with the Yellow Emperor. 

5. The New Palaces and Pleasaunces 
Described and Aff raised 

We pass to Emperor Ming's time, with prosperity in- 
creasing in every direction: so, a new Towtnng-Virtus 
[palace], followed by erection of Solar Virtue. 

Then was opened the special gate [to the palace enclo- 
sure] Southern-Beginning, and the stately structure of 
the Door-of-Fulfilment. 

[ 63 ] 



CHANG HENG 

Benignity and mercy shone on [the gate] Honor-for-the- 
Worthy, the repute of justice was extolled [in the gate] 
Winter-Metal. 

Cloud-dragons were painted on the gate that led to the 
Eastern Road j Spirit-Tigers were depicted in the western 
quarter. 

Built were the twin watch-towers Symbol o£ Majesty, 
emblems of the time-honored standards in the Six Scrip- 
tures. 

Within the enclosure came F/Vz/^j-Embodied, Tower-of- 
Beauty, [the halls] Celestial-Prosperity, Light-Dispens- 
ing, 

Reviving-Decrees, Welcome-of-Spring, Enduring-Peace, 
Lasting-Repose. 

Flying galleries through which the Emperor passed like 
a spirit, unseen by mortal men. 

There in a pleasaunce is Glittering-Dragon Pool and Fragrant 
Grove. . . . There stands the palace Lasting Peace with 
bamboos in winter leaf, with an underground stream . . . 
and various kinds of birds (6 species noted). 

To the south (of the Palace of Solar Virtue) are a terrace 
and two other buildings . . . with rare trees and precious 
fruits "in the charge of the Kou Tun officer." West of it is 
the Shao Hua hill with a kiosk on it. Within the [ancient 
Chou] Nine Dragon building space is what is [now] called 
Felicitous-Virtue. 

Its west and south doors being without ornament or carv- 
ing, this absence of it pleased our Sovereign (Emperor 
Ming), lover of moderation. 

(Thirty li away) to the east is the Great Lake . . . with its 
water fowl, its bulrushes, tortoises, and shellfish. On its west 
side is the Peace-rejoicing campus with its wide views and 
its metal statues of dragon birds in curling flight, and heav- 

[ 64 ] 



FU ON THE EASTERN CAPITAL 

enly horses rearing into the air "cunningly wrought and glit- 
tering in the sun." 

Costly spending has not run to excess, economy has not 
resulted in crudity: the [old] kingly standards have been 
kept, in all the activity of a set goal. 

Behold, here is Li [the Ritual Mind], the right and fit- 
ting in full display : from the beginning no violent haste, 
"the work completed before the day is out." 

As if [the sovereign] had said "The laborers labor, the 

dwellers are at ease." 

So, with recollections of Sage Kings and their "thatched 
cabins," there was construction of the three sets of buildings, 
the Ming T'ang, Pi Yung and Ling T^ai, "where doctrine 
is disseminated and abiding principles are set forth." The 
Ming T^ang consisted of 

duplicated shrines, a series of lodges, with eight windows 
to each of nine compartments. 

[Above] they are circular like the sky, [below] square 
like the earth, and the seasons are followed at successive 
sites. 

A bridge of boats was constructed across the transparent 
pool, its waters deep and wide. 

To the left the Pi Yung was built, to the right the Ling 
T'ai established. The good were promoted, the incapable 
replaced, the worthy recommended, the talented selected. 
The P^ing Hsiang [official] observes [impending] malig- 
nant influences J intercession is made for their removal, 
prayers offered against calamities. 

6. The Emfire and Beyond Doing Homage 

Thus it is, on the first day of the new year the princes 
of the realm assemble from all parts, and with them a crowd 
of officials, also by invitation representatives of border and 
remote countries, "all subjects of the emperor, offering treas- 

[ (>s ] 



CHANG HENG 

ure and bringing ceremonial gifts." Below the audience hall 
is a crowd divided into two companies, in front of them rows 
of graded dignitaries also divided into two companies (left 
and right). The bells and drums are there, the guards on 
the steps, as the imperial carriage comes to the porch, with 
"banners brushing the rainbow." Torches in the hall are 
ablaze, and the thunder of the drums shakes the whole neigh- 
borhood, crash on crash. . . . 

The Emperor dismounts by the eastern robing-room 

. . . dons the Heaven-Communicating crown, grasps the 
jade seal, ties the imperial tassels, girds the Kan-chiang 
sword. With his back to the axe-screen and seated on a 
bamboo mat with silk edging, with jade tables to left 
and right, he faces the audience. 

As the hundred dignitaries enter the hall they are marshalled 
according to rank, and their tribute treasures presented. 

The Son of Heaven with three grades of formal bows 
makes ceremonial acknowledgement. 

How stately, how majestic! 

How correct, how impressive! 

Truly the most impressive spectacle in the world! 

The emperor then invites some kung hou (princes and high 
nobles) and chief ministers to come up to the throne. There 
he confers with them on untoward circumstances and govern- 
ment measures. 

He pities the people in their sufferings and removes their 
woes. 

Should one man not obtain his place [in life], it was as 
if he [the Emperor] had pushed him into a dry moat, 

Bearing the heavy burden of THen hsia [All-under- 
Heaven] having no idle moment for being at rest. 

The emperor opens the granaries with their wealth in reserve 



FU ON THE EASTERN CAPITAL 

and distributes largesse to the officials down to the meanest, 
increases their allowances and gives presents of sacrificial 
animals to ducal families. 

At the feast all drink rich wine and savory meats, "prince 
and subject intoxicated with good will." Then the great com- 
pany disperses in elated mood, but the emperor 

. . . labors at repeated investigations, unceasing in energy. 

His pure repute unites with the Dark Virtus [of the uni- 
verse], his weighty influence is imbued with the self-so- 
ness [of nature]. 

Patterned on the Early Spirits [i.e. the Sages], following 
in their tracks, he is sure to think thrice for fear of trans- 
gression. 

He summons the virtuous from obscurity; is open to re- 
proof in the straightest language. 

He invites some village incorruptible, sending him the 
silk of commission. 

High and low communicate in fellow-feeling — behold, 
there is peace and joy. 

7. The Sacrifice to Heaven 

Turning to the sacrifice to Heaven, the report in the 
suburb of Earth's achievement [with the crops], [the 
Son of Heaven] prays to Heaven for prosperity \ju\^ 
takes thought how to be scrupulously careful [ ? in all he 
does], with every prescribed action complete in [its] 
reverence, every rite perfect in [its] solemnity. 

Thereafter the sacrifice is made in unalloyed sincerity, 
the offerings are presented: as is said, "In very truth a 
Son of Heaven!" 

With robes and sacrificial crown correct in every required 
respect ... a special carriage is used with a special banner 
symbolizing "the undeviating order of the heavenly bodies," 

[ 67 ] 



CHANG HENG 

and six black horses with special ornamented harness j and 
bells, attached to the carriage, chime as it moves. . . . Robes 
and carriage trappings make a harmony of color. 

"Nine-times-nine carriages" (three abreast) follow, with 
their scalloped banners and blue-green canopies j and only 
when the previous carriages are given the command and are 
away, do the next first carriages move off: a color scheme of 
luan banners and tiger-skin covered carriages and silken flags. 
Then come the household cavalry with their standards ( speci- 
fied), a mass of halberds with their horses' manes embellished 
and the men wearing pheasants' tails. 

Flanking blood-horses from the Ch^eng-hua [stable], 
streaming along with tossing trappings. 

Light troops follow . . . and the whole procession moves 
decorously and smoothly in mass formation, the carriages 
squeaking and clanging (su su, hsi ksi, yen yen, ling ling). 

Before the rearguard is out from (under) the city gate- 
tower, banners are wheeling on to the suburban plot. 

[All this] in admiration of Sage-King Yii's supreme 
beauty of [sacrificial robes], in devout reverence for the 
discerning gods! 

The flutes and zithers with the accompaniment of the drums 
discourse six numbers, and when that is finished, the ranks 
of palace dancers perform. 

The great Sacrifice is celebrated, the spirits of the hills 
and rivers all served in due order. 

The flames of the massed fire soar up, higher and higher 
goes the smoke, up to the Supreme One. 

The gods enjoy the odors and take note of the virtue 
[thus displayed], they bless the discerning lord with 
New Year good fortune. 



[ 68 ] 



FU ON THE EASTERN CAPITAL 

8. Other Ritual Duties and Moral 
Allegorizing on Them 

A. The Ming T^ang and the Five Shang Ti 

Thereafter honor is paid to the [Five] Shang Ti in the 
Ming T'angj Kuang Wu is promoted to be associate. 

[Their] directional positions are distinguished aright and 
the procedure rectified, the Five Essences take the lead 
[in turn] and come pressing forward. 

[Special] honor is paid to the Fiery-colored One, the 
four [other] spirits are energetic and indeed sympathetic. 

Thus it is that spring and autumn follow in due course, 
the four seasons come and go. 

B. The Ancestral Sacrifices 

[Lo] the fervently filial heart touched by the [seasonal] 
products, thinking much [of its forebears] ! 

In person the Emperor sees to it that the utensils and meats 
for the four seasonal sacrifices are all in order in the several 
ancestral fanes. The rites are then carried out with "enlight- 
ened decorum" and the Wan dance performed. The spirits 
of the departed emperors descend and feast. Having all been 
well wined, they "send down good fortune in masses." 

C. The Ploughing of the Field of Heaven 

When the day of good omen for the farmers arrives, and 
the pulses in the fertile earth begin to stir, the Emperor 
goes in his luan carriage with azure colored horses, riding 
with a plough by his side. In person he drives three furrows 
in the "heavenly field," preparing the thousand mou (each 
about one tenth of a Western acre). There the millet for the 
sacrifices to his ancestors and to Heaven is grown, so there 
he perfects his thoughtfulness in physical exertion. Thus the 
mass of his subjects is stirred to "be energetic in weeding 
and hoeing." 

[ 69 ] 



CHANG HENG 

D. The Archery Rite and Its 

Preliminaries 

Alike in the spring there is held in the Pi Yung, at sun- 
rise, the ceremony of the contest in archery: a symbol of 
concord. The preparations are duly made beforehand: the 
stand for the bells and drums . . . every instrument in place, 
flags a-flutter, with 

Po I as originator and helper in procedure, Hou K^uei 
seated and doing his task. 

The targets are set up with their five inner and outer 
demarcations, the tweezers in readiness (for plucking out the 
arrows from the targets). 

The imperial carriage at dawn waits by the eastern steps 
until the mists clear and "the sun has risen over Fu Sang." 
Then the Son of Heaven starts the carriage moving and with 
"Ta Ping to gentle his steeds along and Feng Hou as es- 
cort" smoothly approaches the archery building. 

E. The Mystic Arrow: The Significance 

of the Wine Festival 

and Entertainment of the Veterans 

With the rites duly unfolded, with the musical instru- 
ments all there, with the Wang Hsia finished, with the 
Tsou Yu being played, the thumb-ring fitted, the carved 
bow stretched. 

It [i.e. the hitting of the target] stimulates the extra 
grain-shoots in late spring, reveals the heart of integrity 
in a far-reaching parable. 

He [the Emperor] promotes enlightened virtue [te] 
and exalts the imperial heritage j he washes away the 
fierce appetites of ravening greed. 

The spirit [lit. wind] of humanity overflows and streams 
abroad J righteousness is stimulated and speeds into dis- 
tant parts. 

[ 70 ] 



FU ON THE EASTERN CAPITAL 

When sun and moon are met in the Tail of the Dragon 
[tenth month], he pities the pain of labor in his people's 
tasks. 

They all rest from their labors and merriment is increased 
by the spring wine. 

So also he (the Son of Heaven), with his right arm bare, 
cuts up the meat for entertaining the country's veterans, con- 
descending from his high estate to inculcate the spirit of 
respect. Thus by his own strict attention to decorum he 
teaches his people "not to snatch," and the fame of his prin- 
ciples overflows into the heavenly regions. 

9. The hater Han Hunt 

The civil virtus having been revealed, restrained martial 
energy is set forth. 

With three quarters of the year devoted to agriculture, 
[in the fourth quarter] the dazzle of might [is dis- 
played] in China. 

In the second month of winter, a great review is held in 
the Western Park. 

Before that the game-keepers get to work, round up the 
birds and beasts. On completion they report "All ready." 
Then the light carriages are started with four well-trained 
horses to each, "with lances and spears like a forest" and ban- 
ners all flying. The troops arrive and take up their stations, 
awaiting the signals for staying, advancing, and retreating. 
Being fully directed, those in charge kill the sacrificial animal 
as a warning of what will happen to those who disobey, the 
principles and prohibitions being clearly set forth. 

With their torches upheld, the soldiers are like the stars 
in their dispersion, some in goose-crane [wedges], some 
like fishes in files, or fanning out like a spread of wings. 

The carriages drive with reasonable speed j they shoot in a 
straightforward way, and the arrows do not tear the fur. 

[ 71 ] 



CHANG HENG 

Every kind of bird and beast is caught. . . . Three drives are 
enough on the principle of propriety, for (the Son of Heaven) 
"does not exhaust his pleasure, with a view to teaching mod- 
eration, does not exterminate the creatures with a view to 
displaying human-heartedness." He admires (Sage-King) 
Ch'eng T^ang for leaving one side of the net open and in 
his sacrificial prayer showing concern for the common man, 
and takes pattern from King Wen at Wei Yang where he 
"lost a bear but captured a man." 

The dew of his mercy extends to insects and worms. His 
majesty overawes the eight directions. 

He thus being both "truly refined and truly warlike" 

. . . the [Chou] pursuit of the chase in Ao, what a trifle 
it was: [King Ch'eng's] spring hunt in Ch'i Yang, how 
can it be taken as worth reckoning? 

10. The Driving Away of the 'Demons 
of Plague 

And then at the end of the year there is the warding oflF 
of pestilence, the expulsion of a host of demons. 

The Fang Hsiang officer is there, battle-axe in hand, and 
shamans carrying rush brooms. Also a great crowd of boys 
and girls in red turbans and black clothes, with peach-wood 
bows and arrows. These they shoot off aimlessly, like a rain 
of stars, and "malignant demons are sure to all fall dead." 

Like flames they race away, are like falling stars, Ch4h 
I [Red Epidemic] is driven out of the Four Seas. 

Thereafter, the demons having ascended the Milky Way, 
their flying bridge [with earth] is cut. 

The // chu^ are killed, the hsU Uuang [headless goblins] 
are cut in two, the wei i [with head like a wheel] be- 
headed, the fang Hang [marsh goblins] brained. 

^ It is not always clear whether these are Individual or class names. My 
conclusions are based on the commentary, Shan Hat C/iingy Chuang Tzu^ 
ch. 19, and the essay "Defining KueP^ in Wang- Cheung's Lun Heng. 

[ 72 ] 



FU ON THE EASTERN CAPITAL 

Keng Fu is imprisoned in Ch4ng Ling (stream), Nii Pa 
[drought demon] drowned in Shen Huang [stream], 
the K'uei [dragon shaped monster in trees and rocks], 
the hsil [drought demon] and Wang Hsiang [tree and 
rock bogy] destroyed, [the eight brothers] Yeh Chung 
[roving fire demons] exterminated, and Yu Kuang [a 
fire demon] wiped out. 

Because of this the demons all over the world quake with 
fear — how much more the baby goblins and [Old Father] 
Pi Fang [who sets houses on fire]. 

The Tu Shuo [fairy mountain] is a protection, with Yu 
Lei on guard, with Shen T^u to aid him — an interlock- 
ing device, a roped together fence.^ 

With eyes searching every hole and corner, the officer 
[and his lictors] arrest the demons left over: the houses 
in the capital being scrupulously cleansed, nothing is 
there which is wei [contrary to right and to rule], 

II. The Tour of Insfection 

Thus with the Yin and Yang interchanging harmoni- 
ously, with all creatures burgeoning in due time, with 
the oracle having been taken for a felicitous journey, 
[and] the final reply being truly favorable, he [the Son 
of Heaven] ascends his carriage and goes in royal prog- 
ress to Mount T^ai, urging [the people] to the work on 
the land in the Great Plain. 

^ Shan Hat Ching ap. Ltin Heng 65 (the modern text of S/ian Hat 
Ching does not contain this passage) says, "In the middle of the Cold 
Ocean there is the Tu Shuo Mountain, and on it a great peach tree, 
in size 3000 It. The north-east side of its branches is called Demons' 
Door, where 10,000 demons go in and out. Above are two god-men, one 
called Shen T'u, the other called Yii Lei. They control the 10,000 demons, 
and the malignant, injurious demons they bind with cords of rush and feed 
to tigers. Thus it was that the Yellow Emperor made a rite by which 
they could be expelled. He made a statue of a man out of peach wood, 
painted Shen T'u and Yii Lei and the tigers on the door of the house, and 
hung up ropes of rush with which to keep away the demons," 

[ 73 ] 



CHANG HENG 

[As] he unifies weights and measures [everywhere], 
standardizes the wheel tracks, [so] he metes out equal 
treatment to the poor and the well-to-do. 

He examines the shady and reputable officials for de- 
motion and promotion j then he turns round his ban- 
ners and goes back. 

He observes the old burial grounds of former sovereigns, 
musing at length, cherishing the past. 

He waits for the autumn wind, and then he goes west 
[to Ch^ang-an], and there makes faithful sacrifice to the 
High Ancestor [the first Han emperor]. 

As his journeyings coincide with the rousing of the insects 
in spring, so he plans abundant harvests and rewards the in- 
dustry of the field bailiffs. His scope of observation extends 
"from the valley of the rising of the sun" to the dark region 
of K^un Lun in the west, and from there he projects his mind 
to the ends of the heavens and the further days in the future, 
thereby planning the pattern for future ages to follow. So 

S| "he returns home to be free from his labors, bearing in his 

"'^\ breast many blessings along with peace." 

gi All felicitous influences are gathered to perfection, the 

|j (fabulous) isou yU and t^eng huang horses, stabled and "obe- 

I dient to favor," the (fabulous) luan and feng-huang birds 

singing, (fabulous) trees and plants flourishing in the royal 
orchard. 

The repute of the [imperial] grace is a wide-spreading 
coverlet, the dew of it reaches to obscure deserts: in the 
north bringing peace to Ting Ling [30° N. 80° E.: 
Siberia], in the south concord in Yueh Shang [South An- 
nam], in the west including Ta ChSn [Roman Empire], 
in the east going over to Lo Lang [Korea] : to peoples 
whose languages must pass through nine [different] in- 
terpreters, all bowing their heads and doing homage. 

[ 74 ] 



fn 



FU ON THE EASTERN CAPITAL 

12. The Peofle^s Argument With 
the Throne 

Hence, if there be discussion over removal of the city 
and changing the capital, then it should be in accordance 
with the precedent of P^an [Keng's removal to] Yin: 
[if discussion on] change from luxury to frugality, then 
it [should be] in agreement with the [simple] beauty 
[of the buildings] in the Ssu Kan [ode] : [if discussion 
on] an ascent [of Mount T^ai] for the feng [sacrifice] 
and descent [to Liang Fu] for the shan [sacrifice], then 
the [Han] virtus [should be] united with [the virtus 
of] the Yellow Emperor. 

With action being non-action and busy-ness not being 
busy, there is the lasting [allegiance of] the people bring- 
ing all-pervading peace. 

He [the Emperor] observes moderation and economy j 
he esteems untrammeled simplicity. 

He muses on Confucius' "self -subduing" j he treads in 
Lao Tzu's abiding sufficiency. 

Thus he can cause his mind not to be deflected from the 
right course, his eyes not to see what can arouse desire. 

He despises rhinoceros horn and ivory, looks askance at 
pearls and jadej he lets gold lie buried in the mountain, 
thrusts aside rare jade in the gorges. 

Cock-kingfisher's feathers he does not tear offj tortoise 
shells he does not collect. 

What he honors is solid worth j what he prizes is grain. 

[Thus] the people discard the branches [non-essentials] 
and go back to the root [essentials], cherish loyalty and 
embrace sincerity. 

At this very time all within the [four] seas join in re- 
joicing, exclaiming, "The virtus of the Han emperors, 
how excellent is its scent!" 

[ 75 ] 



CHANG HENG 

The ming chieh^ [a fabulous plant] is dijfficult to grow, 
and therefore in a neglectful age it is not to be seen. 

Only our sovereign lord can plant it, using supreme har- 
mony and peace, [only he] will be able to count [its 
leaves] on the palace steps. 

That being so, then how can the Tao [Way] not cherish, 
how can the [imperial] transforming power not soothe? 

Fame flies along with the winds [of Heaven], mercy 
follows the movements of the rain-clouds [of Heaven]. 

[The Emperor says] "It is We on whom all creatures 
depend — what else is there for them to seek!" 

"The virtus covers [all], as the heavens cover [all], a 
flame that blazes, a light that shines. 

"[We] count the small house [i.e. institutions] of the 
Three Kings [the Hsia, Shang, and Chou regimes] too 
meagre a dwelling, [We] overtake the Five Emperors 
in their long and urgent course. 

"[We] tread in the far-off steps of the Two August Ones 
[Fu Hsi and Shen Nung] — who says [We] have started 



iCl 

ai too late and cannot overtake them?" 



13. Sundry Reflections on Emferors and Their Way 
i' of Life 

I have not finished cataloguing the excellencies of the 
Eastern Capital. Since I happen to be suffering from a 
stupid complaint, I am unable to assess such subtle es- 
sences. 

Hence my reply [Sir] is crude: in rough outline it is as 
follows. 

Suppose a man slipping into dissipation and forgetting to 

* The ming chiehh leaves were supposed to open on the first day month 
by month. Hence the Emperor had an infallible guide for the longer and 
shorter months in the year and could keep the calendar right. 

[ 76 ] 



FU ON THE EASTERN CAPITAL 

turn backj he would be giving rein to his mind without 
being aware. 

[Suppose] him enjoying himself without any restraint, 
later he would be entangled in misery. 

"One word can come near to destroying a country" — 
it is not I who have learnt it. 

[As the proverb has it], the man with expertise in carry- 
ing pitchers holds on to it and does not lend his vessels 
out. 

How much less can the inheritor of the imperial patri- 
mony treat his heaven-given station lightly! 

Look up to the Two Ancestors [Emperors Kao and 
Kuang Wu] and their immeasurable tasks j their upward 
flight was constantly attended by fear of danger, as if 
a man rode at a gallop and had no bridle. 

[Take the story] of the white dragon who assumed the 
guise of a fish and was roughly handled by Yu Chu. 

[So] [the lord of] a myriad chariots, although he has 
nothing to fear, may yet be intimidated by a common 
man. 

All day he does not leave his protective entourage [bag- 
gage wagons] '^ — if he makes incognito excursions, where 
does he go? 

Monarchs stop their ears with [voluminous] yellow 
robes J they do not gaze abroad from within their car- 
riages. 

Girdle pendants control their demeanor j carriage bells 
cause measured [pacing on] the road. 

In walking [the chiming] of the jades does not change, 
in driving [the horses] do not break step. 

'^Allusion to Tao Te Ching^ 26: "A man of consequence though he 
travels all day will not let himself be separated from his bag-g-age-wag- 
gon.» (Waley) 

[ 77 ] 



CHANG HENG 

[He] turns back the fleet horses to draw dung-carts, not 
even begrudging such steeds as Yao Nao and Fei T'u. 

14. Plain Facts of Social Economy 

When he [the Emperor] proceeds to use up the [na- 
tion's] wealth and avail himself of its goods, there is the 
constant fear lest the [different] classes of commodities 
be all used up: [when] he taxes and uses the corvee^ 
there is the constant fear lest men's strength be ex- 
hausted. 

Let [the goods] be allocated according to principle, [the 
labor] be used at the proper times. 

On the hills let there be no [indiscriminate] felling and 
lopping, in hunting no killing of [life in] the womb. 

Then the plants and trees will flourish, the birds and 
beasts will multiply. 

The people will forget their toil, will be happy to con- 
tribute their wealth [to the nation's use]. 

I So the hundred clans will be alike in having abundance j 

gi high and low share this peace and amity. 

li^l The abundant compassion [of Han rule] has long been 

41 stored up, men's hearts [long] been solidly knit together 

B [in loyalty to the Throne]. 

!'■■*, They hold fast to fair-dealing and pay regard to their 

Lord, ordinary people cherish integrity and restraint. 

15. The Nobleman Forced to Face 
Unfleasant Facts 

Enraged by the treacherous violation of Heaven's man- 
date, resentful at the severance of the imperial line 

Whereby Wang Mang, having launched his dark machi- 
nations and conducted them in secret, succeeded for eight- 
een years in his treason, 

[ 78 ] 



FU ON THE EASTERN CAPITAL 

The Emperor ascended the steps of the throne, and the 
blessings of Han rule were displayed in due order. 

With such a consummation we rightly rejoice in the 
royal heritage. 

You, Noble Sir, are now irresponsible, wanting to ex- 
haust the people in the service of [your] pleasures — you 
forget that their hatred becomes a mass vendetta. 

You would exhaust commodities in inflating your arro- 
gance — suddenly the underdog rebels and creates misery. 

It is by water that a boat is borne up, by water also that 
it is capsized. 

Solid ice is forming when frost is on one's shoes j an 
eight-foot tree springs from planting a sapling. 

If [you] rise at dawn [and] become distinguished, 
[your] descendants may still be lazy: how much more 
for him who at first cuts [his] robes on a lordly scale 
is it impossible to change [their] cut [to an economical 
size]. 

1 6. The Final Word from Teacher 
Anonymous 

When Ssu-ma Hsiang-ju and Yang Hsiung made their 
poems on the Ch^ang-an hunts they envisaged, Ssu-ma Hsiang- 
ju the removal of the park walls and surrounding ditches, 
Yang Hsiung a ban on nets and traps. But even if these 
measures had been taken, 

... in the end they would have been unavailing in the 
existing conditions, they would only have revealed the 
defects and excesses [of the regime]. 

Ministers increased their extravagance and overrode their 
sovereign, forgetting the enduring foundation of ruling 
a country. 

Hence, while at the Han Valley [Pass] in the east, 
[Wang Mang's troops] were beating their drums, in the 

[ 79 ] 



CHANG HENG 

west the seat of government was crashing down with no 
hand raised [to save it]. 

It is human for the mind to think that what it has learnt 
is right, for the limbs to be comfortable in habitual pos- 
tures. 

A person in a dried-fish store is not conscious of its smell, 
he being used to what assailed him when he first went in. 

The [Yellow Emperor's and Yao's] Hsien Ch^ih music 
is not in tune with the croaking of frogs [i.e. lascivious 
music] — but some undiscriminating commoners would 
have doubts [as to which is better]. Of those who have 
no doubts, is Tzu Yeh [the great Chou music master] 
the only one? 

17. The Nobleman Refents 

The Nobleman, intoxicated with high doctrine, satiated 
with civilized principles, stimulated to virtue and scared 
of rebuke, with pleasure and fear at war [within him] : 
his wits were astray like a drunkard, day and night jaded 
and weary, like one robbed of his vital essence [ch^i], 
stripped of his fleshly soul [f'o]. 

He forgot what he had argued about, lost what he had 
bragged about. 

After a lengthy pause he said: 

What a vulgar fellow I am, having learnt everything 
wrong and persisted in error! 

Happily I have you, my Master, to point out the true 
south [of doctrine]. 

What your servant has heard is specious and not accord- 
ing to fact J what you. Sir, have said is to be believed on 
the evidence. 

Vulgar fellow that I am, and lacking in discernment, 
from now on I know that the scent of the virlus of Great 
Han is all contained here. 

[ 80 ] 



FU ON THE EASTERN CAPITAL 

Formerly I often deplored the loss of the Three Mounds 
and Five Codes [ancient scriptures]. 

Looking back into past history, I could not descry the 
beauty of Yen Ti or Ti K'uei [mythical emperors]. 

Now that I have heard the overflow [from your] learn- 
ing, Sir, how could Ta Tung's [a Taoist worthy, pre-Fu 
Hsi] [virtus] surpass the present one? 

Your messenger, although unintelligent, has more or less 
come to understand this. 



[ 8i ] 



Chapter VII 



CRITIQUE OF PAN KU'S ''FU ON THE 
WESTERN CAPITAL" 

<><><><xX><X><X><^^c><>0<^^ 

Before coming to a detailed critique of the fu it is necessary 
to diverge for a moment into the field of the literary critic, 
both generally and in relation to the translations as trans- 
lations. 

Consider first the dramatization effected by Pan Ku's put- 
ting his story into the mouths of two imaginary characters, 
the Western Capital Guest and the Eastern Capital Hostj 
and by Chang Heng's putting his story into the mouths of 
a nobleman and a hsien-sheng (teacher), the one old enough 
to have amassed considerable historical knowledge but not 
having the mature wisdom of the other. The question is: 
why this dramatization? In Pan Ku's case the answer is to 
be found in the concluding remarks to his preface. Since his 
purpose — the ostensible one at least — was to stop people 
grumbling over the change of the capitals, this kind of dram- 
atization was a very effective device by which he could 
enable the supporters of the Eastern Capital to score a dialec- 
tical victory over their opponents. This he does to his own 
satisfaction, with almost too crushing effect. Amusing as the 
discomfiture of the Western Capital man is, and as Pan 
Ku obviously meant it to be, yet as drama it is on a very 
simple level. But it is not quite as simple as it looks to the 
modern mind, versed in the achievements of the dramatic 
poets and philosophers of long agesj for Pan Ku had in 
mind the poets of the first century b.c, who had originated 
this device for objectifying their subject matter, and had 
done so with the intent to sharpen the ironical edge to their 
tongues. On the other hand, irony and satire are not entirely 
lacking in simple literature: witness the Chou folk songs in 

[ 82 ] 



PAN KU ON THE WESTERN CAPITAL 

the "Odes." In Pan Ku's case his use of the device is some- 
times lacking in subtlety. He makes his point with so bludg- 
eoning an effect, that one is compelled to wonder whether 
there is something behind it. At once the question springs 
to mind: was the question of the capital so live an issue at 
the time when Pan Ku wrote his two poems as he makes it 
out to be? That time was fifty to sixty years after the event, 
and it is open to doubt whether any but a few old Western 
scholars were still grumbling over the change. We may sur- 
mise that as a matter of fact the lively presentation of the 
argument was an excellent means of belauding the existent 
regime — whether in flattery or from honest convictions must 
remain to be seen. 

With Chang Heng it was a very different matter. His 
bitingly ironical reference to the upper and wealthy classes 
demonstrate this beyond a doubt. Consequently his two in- 
terlocutors, Nobleman Stayed-on-Vanity and Teacher Anony- 
mous, are much more subtle characterizations, more in the 
true vein of Ssu-ma Hsiang-ju and Yang Hsiung. Chang 
Heng's objective is clear j and if in one sense he was imitating 
his predecessor, he did so with the purpose of showing up 
the dangerous superficiality of Pan Ku's much-acclaimed pro- 
duction. For Chang Heng the really important thing was 
that "what is sauce for the goose is sauce for the gander," 
or, for that matter, "people who live in glass houses should 
not throw stones." If Great Han in its former manifestation 
was open to dire criticism as being half barbarous, this later 
Great Han in its present manifestation must be judged by 
the same yardstick: admitting that the later rulers have 
shown themselves less barbarous than the former, does it 
follow that the present regime is really civilized? 

Thus by comparing the two authors' use of the literary 
device of dramatic personalization we not only achieve some 
measure of insight into their respective standpoints and re- 
spective sincerity of mind, we also have a vivid illustration 
of the way in which literature works inside a culture. It may 

[ 83 ] 



CRITIQUE 

be a simple reflection but it is a far-reaching one, that Author 
A having produced a striking work of literary art, Author B, 
of the same generation or a later one, reads it and is gal- 
vanized into following the lead given and striving to go one 
better on it. Admiration leads to criticism, criticism to emu- 
lation in the deepest sense. Only so can cultures remain alive. 
In the case before us there is also this to be borne in mind: 
there was in the Later Han period the clearest consciousness 
among scholars that, as the ancient "Book of Odes" {Shih 
Ching) revealed the ethos of that less developed society, so 
the poetical writings of every age, and more particularly their 
own, were a touchstone of its real ethos in all its heights and 
depths. In a word — and using a Han image of speech — lit- 
erature, that is imaginative literature, is the mirror of its age. 
From this point we turn to the problem which may well 
be in the mind of the reader after struggling with these four 
prose poems. Is this kind of writing really poetry? Perhaps 
not, in the specific modern Western sense. The treatment of 
the subject is conducted at so many points with such prosaic 
attention to detail, with such frank and careful attention to 
argumentative effect, that the works in question can only 
be regarded as prose — rhetorical prose, of course, but none 
the less prose. Yet, can such a judgement be right? Compare 
Euripides' plays, Virgil's Aeneid^ not to speak of Shakes- 
peare's plays and Milton's Paradise Lost and Paradise Re- 
gained. It is here that the reader must go carefully because 
of the unavoidable defects of a translation. The translations 
of actual passages given above have been made with the pri- 
mary object of achieving accuracy of meaning and clarity of 
impression. For the sake of this object the brevity of those 
strictly-ordered sentences in the original has perforce again 
and again been sacrificed. So also has the rhyme with all that 
rhyme creates in the way of crispness and pungency of com- 
munication. So also has the sensuous alternation of alliterative 
sounds and the skilful interchange of longer and shorter sen- 
tences. The only way in which the non-Chinese reader can 

[ 84 ] 



PAN KU ON THE WESTERN CAPITAL 

get some inkling of these essentially poetic effects in the orig- 
inal is by looking carefully at the more succinctly and suc- 
cessfully rhythmic couplets which yet contain some bracketed 
words added. Having ascertained the meaning, then read 
the sentences envisaging those bracketed words left out, and 
so realize how rigorous an attention the author paid to form, 
and how this fu style entailed the achievement of such ring- 
ingly apt language that the scholarly reader of the time 
could not fail to catch the fine shades of meaning without 
the pedestrian help of connectives and the like. But — if that 
was achieved to any substantial degree — then what is pro- 
duced is poetry. There is no need to labor the obvious by 
emphasizing our two authors' command of poetic imagery. 

Finally, on the point of poetry versus prose. A ju poem 
was primarily descriptive of an object {wu)^ either a physical 
object, as for instance a capital city, or a non-physical object, 
such as a state of mind. The description of a physical object 
may take the ju writer over into the spiritual realm, as hap- 
pens in our four ju^ and equally a description of a non-physi- 
cal object may entail much detail as to the physical concomi- 
tants — as happens in Chang Heng's fu on Thought in its 
transcendent phases, the setting to which is an itemized jour- 
ney round the world and under and over it to the verge of 
ultimate space. Being descriptive, these long fu poems furnish 
a large quantity of miscellaneous information, and thus the 
authors had to deal with the problem of arrangement and 
orderly presentation. In the later years of the Former Han 
dynasty, Yang Hsiung showed himself a master hand at 
this, so the tradition and technique of marking paragraphs 
was there by Pan Ku's time. But no one had attempted de- 
scription on so colossal a scale as he did in his account of the 
Capitals and the distinctive practices that went on there. That 
he succeeded in the opinion of his generation is evidenced by 
the applause which these two poems elicited. 

To the modern reader both Pan Ku's and Chang Heng's 
productions constitute a problem, one to which the above 

[ 85 ] 



CRITIQUE 

combination of direct translation and summarization gives 
rather haphazard insight. Arrangement clearly there was, but 
the movement o£ the authors' minds from one phase of their 
subject to another is discerned by the use of more than one 
device. One was rhyme: the final rhyme of the previous 
section would be changed — ^but then the rhyme may change 
every few couplets throughout a section. Another device 
was change of rhythmic tempo, e.g. from a four-word sen- 
tence structure to a six- word onej or the final sentence of 
one section, or the first sentence of the next, might be ir- 
regular in length. Or again there might be an introductory 
connective phrase such as yil tz^u chih shih ("at such a time") 
or erh nai ("thus it was") — although such phrases can be 
found in the middle of a paragraph lending emphasis to 
a particular point. One way and another, even to the stum- 
bling non-Chinese reader, there is a guide to paragraphing. 
The only criticism that can be made is that these fu are lack- 
ing in arrangement on the grand architectonic scale, namely 
with a plain differentiation of major and minor sub-para- 
graphs. On the whole Pan Ku with his simpler and more 
superficial objective made a better job of proportionate ar- 
rangement. Chang Heng's objective was more complicated, 
and this is reflected in his craftsmanship. 

We are now more prepared to examine what each man 
wrote, first about Ch^ang-an and then about Loyang. The 
order of examination is, naturally, to extract the gist of each 
author's poem, then make critiques, and then compare those 
critiques, taking the two Ch^ang-an poems first and the two 
Loyang ones afterwards. 

The Preface to Pan Ku^s "Fu on the 
Western CafitaP^ 

The practice of writing a preface to a fu seems to have 
begun with the philosophically-minded Yang Hsiung (53 
B.c.-A.D. 18). It is a striking feature of the new northern 
school of fu writers, and explains the circumstances in which 

[ 86 ] 



PAN KU ON THE WESTERN CAPITAL 

the poem was written. Thus a preface puts the reader en 
raffort with the author's mood and enables the author to 
concentrate his attention from the very start on the matter 
in hand. This preface here is, therefore, to be taken as self- 
revelatory — at any rate up to a certain point. 

The words "A^o yileh^'^ ("one opinion is") are redolent 
of the contemporary controversial style. On the evidence 
of the Po Hu T^ung and the Lun Heng, the use of this 
phrase implies that there are one or more other opinions 
on the subject in question. Pan Ku does not mention any, but 
his vivid account of the mid-Han literary excitement affords 
a clue. To the bright young court litterateurs of Pan Ku's 
day it might easily seem that the new style which had surged 
into existence in the previous century was a new creation 
by Ssu-ma Hsiang-ju and those others mentioned in this 
preface. Pan Ku, the historian far excellence^ knew better 
than that, as is demonstrated by the postscript to the four 
classes of fu itemized in Liu Hsin's Catalogue (see Han Shu^ 
ch. 30), which draws special attention to the innovations made 
by Ch^ii Yiian, the mid- Yangtze Valley poet and by Hsiin 
Ch^ing, the northern scholar who had taken refuge in the 
south before the Ch4n regime emerged. 

Concerned though he was for the understanding of history. 
Pan Ku's main interest here is that literary new birth of the 
previous century. To do full justice to his statements it would 
be necessary to take each of the writers he mentions — Ssu-ma 
Hsiang-ju, Yli-ch^u Shou-wang, Tung-fang Shuo, Mei Kao 
(interesting that, for he seems to have been a quite frivolous 
kind of writer), Wang Pao, and Liu Hsiang — and mark the 
wide range of new subjects which attracted their interest and 
the new powers of language which they exercised. Most of 
these men have writings extant today, as also have the high 
court officials Ni K^uan, K^ung Tsang, Tung Chung-shu, Liu 
Te, and Hsiao Wang-Chih, mentioned as occasional com- 
posers. But such a study would carry this book too far afield. 

[ 87 ] 



CRITIQUE 

It must suffice to elucidate the special more general features 
of Pan Ku's account. 

It must be noted here that the "Awo yileh?^ opinion cited 
by Pan Ku is found in Mao Chiang's "Preface to the Odes," 
as also the fact that Mao Chiang (second century b.c.) and 
his treatment of the "Odes" was, both in the first century b.c. 
and in Pan Ku's day, not popular with the dominant schools 
of Scripture exegesis. Was Pan Ku standing up for him by 
citing his opinion? 

A second feature is Pan Ku's use of the term ^^yen yU shih 
ts^ung chih cUen^'^ servants of the emperor whose service 
consisted in exercising their powers of language. That jabs 
one's mind. Since when had there been men of this kind 
holding salaried posts at court? In one sense the Ju of the 
old Chou feudal courts had been such, but they were mostly 
scribes, technicians of the pen, in what was a much less liter- 
ate age. There had been, of course, the Chi Hsia and other 
groups of philosophizers, living on the charity of ambitious 
princes. Pan Ku was not forgetting them, for he pays tribute 
to the glories of the literature of the "Three Dynasties" 
(Hsia, Yin, and Chou). It is difficult to imagine Pan Ku 
having any other meaning in his mind than that from the 
time of Emperor Wu on there was a new and unprecedented 
phenomenon, a set of salaried men at court whose real busi- 
ness and main-time job was creative writing. The study of 
the biography of Ssu-ma Hsiang-ju and that of Yang Hsiung 
in the "History of the Former Han Dynasty" reveals this 
clearly enough. They were not expected to be administrators 
but to be poets. 

A third feature is the striking emphasis laid on there hav- 
ing been two kinds of creative writing. One, to be sure, was 
singing the praises of the imperial virtus-^ and one would like 
to know what Pan Ku meant by his "and thus they perfected 
their loyalty and filial piety." It looks suspiciously like a ref- 
erence to a sense of high duty driving a man to soft-pedal 
the shortcomings of the throne. However, the other kind of 

[ 88 ] 



PAN KU ON THE WESTERN CAPITAL 

composer — and Pan Ku mentions them first — interpreted 
popular feeling on current affairs "and so communicated sa- 
tirical remarks." Ssu-ma Hsiang-ju's fu poems in the Wen 
Hsiian are admirable examples of this class, and one deli- 
cious illustration comes in Yang Hsiung's Kan CUilan Fu 
where he describes, in the preface, how he was ordered to 
write up the special sacrifice at the Kan Chilian palace, and 
how a "wind of irony" caught him as he did so. 

A fourth feature is Pan Ku's comparison of the new literary- 
creations with the old as found in the sacred Scriptures. In 
point of value the former are to him just as important as 
the latter. Thus in that Scripture-worshipping age of Later 
Han, here is one writer frankly comparing antiquity and re- 
cent times on an equal footing. His language of comparison 
is a masterpiece of neat aptness, clarity, and conciseness. The 
old literature was ]u fi^ "like that," the new is ju tz^u^ "like 
this." The contrast of a "that" out there in the distance with 
a "this" intimately known in one's self, is a feature of the 
Taoist philosopher Chuang Tzu's terminology, and in the 
Tao Te Ching (ch. 12) stand the words "discard the that and 
lay hold of the this." 

In conclusion, it is only in his last words that Pan Ku 
reveals his motive — or at any rate his ostensible motive — in 
contrasting the two regimes. Since he gives this as the con- 
founding of certain Ch^ang-an grumblers, why then all the 
detail about the creative writers? There are two questions 
which naturally arise. One is whether after fifty years of 
success in Loyang there could have been any serious agita- 
tion for return to Ch^ang-an. The other is whether the 
markedly controversial and theological temper of the schol- 
arly world was not tending to dry up the fountain of cre- 
ative writing. My conjecture is that by the double emphasis 
in this preface Pan Ku wished to commend his composition 
to Emperor Chang and his advisers as thoroughly loyal in 
temper, and at the same time urge the importance of fos- 
tering poetic talent. I can find no evidence of recognized 

[ 89 ] 



CRITIQUE 

buildings in the Loyang court where the budding masters 
of language were encouraged to congregate. 

I. Introduction of the Dramatis Personae 

As the reader gets further and further into each of the 
four fuy he will find himself questioning at certain points 
whether the sentiment expressed is really representative of 
Interlocutor A or B speaking in frofria fersona^ or whether 
Pan Ku or Chang Heng have more or less forgotten that 
they were not speaking in the first person. The question that 
thus arises in the field of literary criticism does not concern 
us here, but, assuming that the disguise does in places wear 
very thin, the quality of self-revelation thus made does con- 
cern us. Certain contrasts between the personality of the two 
poets may thus emerge to view. 

2. The Tofografhy and Geografhy of the Site 

The case for Ch'ang-an starts off with the argument for 
it as the premier site for the seat of government, possessing 
as it does unique natural advantages. The ordered recital 
of the mountains and rivers is impressive, for it is a com- 
pletely naturalistic statement — one seems almost warranted 
in saying, a severely scientific approach to the subject. Moun- 
tains with fortifiable passes for protection, an abundant water 
supply, fertile soil, these are the factors stated as influencing 
the royal decision. Add to this the details given in Section 4: 
the "five grains" (presumably including rice) with the help 
of irrigation producing abundant crops and a canal (or series 
of canals) linking the Wei River with the Yellow River and 
so establishing boat communication with the sea. Pan Ku's 
purview stretches from the China Sea to the upper reaches 
of the Yellow River in the west (about meridian 95°). That 
is the scale on which he views the setting to the Ch^ang-an 
site, and if his mundane naturalism seems to be lessened by 
his reference to the Chou regime as ruling by the virtus of 
the dragon, it is noteworthy that in the next clause the more 

[ 90 ] 



PAN KU ON THE WESTERN CAPITAL 

recent regime of Ch4n is symbolized by the realistic emblem 
of a tiger. We descry here, therefore, a highly naturalistic 
mind at work, one with quite acute appreciation of geo- 
graphical features. The section is a masterpiece of orderly 
summarizing. 

3. The City and Its Suburbs 

Quite clearly Ch'ang-an was what we call a metropolis: 
a small one (according to modern ideas) but none the less 
a metropolis. The description of the walls and moats and 
the crowded streets may strike us as exaggerated, but there 
is no mistaking the verisimilitude attaching to the picture as 
a whole. It is a very striking picture, and to the historian 
a very valuable one, for, although Pan Ku may never have 
been to Ch'ang-an himself, his father had — and possibly 
choked over the dust clouds in the streets. The glimpse we 
get of well-to-do bourgeois shopkeepers introduces us to a 
social phenomenon of Han times, namely flourishing retail 
trade with its concomitants, trade in luxury goods, large scale 
manufacturing, in some lines at any rate, and the power of 
trade-produced money. The question arises: how far is this 
mercantile Ch^ang-an comparable to the Rome of the Caesars 
or Paris of the late Middle Ages, or even of Louis Le Grand's 
time? Chang Heng's picture will add to our knowledge, but 
for the moment the relevant document for consultation is 
Pan Ku's "History of the Former Han Dynasty" (ch. 24), 
the "Monograph on Food and Money" of which Dr. Nancy 
Lee Swann has made so copious a study.^ There we find ample 
confirmation of the picture drawn here, and this from the 
point of view not merely of the capital but also of the prov- 
inces. Particularly clear is the continued plebeian status of 
the great Han capitalists. They had influence in all sorts of 
ways, within the law and without, but the Han social order 
remained essentially a patrician-plebeian one. This is patent 
also in the Western man's description of the further suburbs 

^ Nancy Lee Swann, Food and Money in Ancient China ^ Princeton 
University Press, 1950. 

[ 91 ] 



CRITIQUE 

where the aristocrats and captains of industry had their homes. 
They did not live in the crowded streets o£ the city. The in- 
formation given is particularly useful because Pan Ku stresses 
the fact that it was imperial policy to have them there. 
Through much of Former Han times the struggle went on 
on the part of the provincial magnates, the scions of the 
imperial house and the great families plotting in the hope 
of partial or complete independence. Hence the skilful device 
of calling on potential trouble-makers to honor themselves 
by guarding the High Ancestor's tomb. 

The final emphasis must lie on the fact that never before 
had the country had such a metropolis. Clearly, the story 
of it captured Pan Ku's imagination. 

4. The Ifnferial Demesne 

The area under scrutiny is given under the term used in 
Chou feudal times, the "imperial demesne." In some sense 
the idea of a privileged area where the royal beneficence 
could most make itself active was still a real thing 5 a con- 
tented peasantry in localities so near to the imperial resi- 
dence was plainly of importance. The nature of the crops 
is noteworthy, not only cereals but mulberry trees for silk 
manufacture and hemp for the ordinary people's clothing. 
The narrative does not overtly indicate that the high pro- 
ductive power of the region was by no means the work of 
unaided Nature. For that we have to go to Pan Ku's his- 
torical work "History of the Former Han Dynasty" (ch. 29). 
There we see how the establishment of a great capital con- 
stituted a problem of food supply, one with which the Ch'in 
Emperor had tried to deal. It was not, however, till Emperor 
Wu's reign that the problem was adequately tackled. There 
was an enormous expenditure of forced labor on large scale 
irrigation, for the work was of an experimental kind, and 
where one experiment failed, alternative measures were tried. 
The construction of canals, useful for transport as well as 
for irrigation, is an indication of the ultimate success at- 

[ 92 ] 



PAN KU ON THE WESTERN CAPITAL 

tained. In this way came a considerable advance in the tech- 
niques of water engineering. May we not say, with all due 
respect to the legends of Sage-Emperor Yii, that here for 
the first time in China's history a powerful and vigorous 
central government had had the resources and the engineer- 
ing intelligence sufficient for accomplishing a large-scale agri- 
cultural miracle? 

With regard to the exotic birds and beasts native to such 
very distant countries, the rational imagination boggles over 
the problem of transportation. For example, for so bulky 
an animal as a rhinoceros to be brought from the Malay 
Peninsula must have required months of sea passage and a 
successful solution of the problem of feeding en route. Is, 
then, Pan Ku by any chance exaggerating wildly? On the 
other hand, he is so very precise in his details of location, and 
the existence of a zoological enclosure in the West Suburb 
is well-enough attested. (See Section 9 on the Imperial Hunt). 

5. The Emferor^s Personal Dwelling-Places 

Pan Ku has been so good thus far with his precise, if some- 
what impressionistic, descriptions that, now that he reaches 
the palace buildings, it is disappointing to find how sketchy 
he is. We get no conception of the size of the main building, 
nor how it stood in relation to the subsidiary buildings round 
it. There is ground for the surmise that Pan Ku got tired 
of his prolonged task with the "History of the Former Han 
Dynasty," involving as it did endless attention to detail, and 
that the writing of this fu and the one on Loyang was a relief. 
That being so, we can understand his being a littly sketchy 
here. But there may have been a lack of detailed information 
in the imperial archives at Loyang: most of Ch^ang-an went 
up in flames when the Wang Mang regime collapsed. 

One thing is noticeable. There is no mention here or in 
Chang Heng of any building for the service of religion. 
Apparently there were no specifically religious rites carried 
out in the palace enclosure. All that was relegated to places 

[ 93 ] 



CRITIQUE 

outside the capital, and the question is whether some old, 
old taboo was not operative behind this arrangement. On 
the other hand, there is a categorical affirmation of what 
appears in various passages in the Canonical Scriptures, namely 
that the ordering of life at the imperial level should con- 
form to, and thus symbolize, the order of Heaven and Earth 
(viewed as two correlated and correlating mystic entities). 
The statements on this matter with regard to Chou institu- 
tions are necessarily subject to some suspicion as denoting 
an ideal rather than an actual state of affairs. In the light 
of Pan Ku's statement, however, we clearly have to take 
into account the probability that the earlier Han, and yet 
more the later Han dynasty, made serious attempts to carry 
the symbolism into effect. That would seem to bring religion 
right into the palace. 

With regard to those cross-beams and pillars, etc., etc., 
they belonged to the building that the first Han emperor's 
chief minister erected for him (end of the third century b.c). 
As all educated people knew in Pan Ku's age, the Emperor 
had been annoyed at the extravagant expenditure entailed, 
but abated his anger when the minister explained to him 
that as a matter of royal prestige it was necessary for him 
to impress his subjects with the pomp and magnificence of 
his court. The Western Guest has nothing to say on that 
point, but his creator clearly wanted to bring home that the 
building had been constructed regardless of expense. With 
regard to the whole set-up he makes his point. In true fu 
style, he specifies nine of the buildings by name and says 
there were more which "it is impossible to discuss in detail," 
and then goes on to a shower of descriptive expressions, "with 
solid foundations, soaring into the air, above and below a 
blaze of color: a medley of shapes cunningly contrived, to 
the eye each different from the other. . . ." We stumble on 
the question which must occupy our attention later: did those 
dragons' pinions on the cross-beams only have reference to 
old animistic beliefs, or were the royal craftsmen spurred 

[ 94 ] 



PAN KU ON THE WESTERN CAPITAL 

on to new flights of imaginative workmanship, seeking to 
make beauty for beauty's sake? 

6. The Rear Palaces 

The information we get here is of the kind which does 
not appear in the same way in the sober works of the period j 
and yet the influence of the harem is discernible on most 
pages of Han history. The large number of buildings is, as 
Pan Ku points out, indicative of the numbers employed there. 
These ranged from expert ladies' maids to the meanest scrub 
woman, and in that connection Pan Ku notes one character- 
istic of life there : the fortunes of the inmates were constantly 
changing. There was no bar to a beautiful girl of humble 
parentage being employed and finding favor with her mis- 
tress, becoming a lady-in-waiting and so maybe catching the 
imperial fancy. Europe had the same experience in much 
the same form. So also this picture of "Corinthian luxury" 
is familiar to the historian of the west as well as the East. 

7. The Administration Offices^ the Archives ^ 
and Other Buildings 

Having devoted about 80 lines to his rococo description 
of the housing of the emperor's personal menage. Pan Ku 
disposes of the rest of the palace hill in 26 lines. It is a mar- 
vel of compression, disappointingly so when we think of the 
information we should like on the oflice structure of the 
various ministries. However, what detail we do get in the 
whole passage is of exceptional value. 

The first part is in quite conventional language after the 
pattern of eulogies. Yet even there we find a nugget of gold, 
the reference to the popular song made in the capital. It 
strengthens the impression which H. H. Dubs' History of 
the Former Han Dynasty fosters, namely that a rule of law 
and the rudiments of a constitution began to emerge. Pan 
Ku would have us believe that this was appreciated whole- 
heartedly by the citizens of Ch'ang-an. The emergence of 

[ 95 ] 



CRITIQUE 

a constitution, tentative and uncoded though it was, is a fact 
of major importance in the history of the Han era. 

Coming to the four specified buildings (which were to the 
north of the Wei Yang Palace), we are introduced first to 
that epoch-making innovation of the last decades of Former 
Han, a state library, in which a collection of manuscripts of 
all kinds was reduced to order, classified and catalogued (see 
"History of the Former Han Dynasty," ch. 30). To what 
extent the governmental records were stored in the T4en 
Lu and Shih Ch'u buildings does not appear, but the refer- 
ence to specially-commissioned officials setting forth the teach- 
ings of precedent points in the direction of the more im- 
portant documents being stored there under lock and key. 
Pan Ku's picture is that of a repository to which only a select 
few were given access. Among these, it is natural to assume, 
the Erudites, should be numbered, they being the chief offi- 
cial exegetes. Actually "^0 shih^'^ ("Erudites") does not occur 
in Pan Ku's phrasing here, and in any case the Erudites did 
their teaching at the Ta Hsueh, the university which Em- 
peror Wu inaugurated, the buildings of which were two miles 
from the city. However, since the conference of Erudites 
and other famous scholars in 51 b.c. held its sessions in the 
Shih Ch'ii Pavilion, it is clear that Pan Ku had this confer- 
ence in mind as he wrote. That being so, the student can re- 
fresh his memory of the proceedings by reading over Tjan 
Tjoe-som's admirable Po Hu T^ung (I, 89-94 and passim j 
Leiden, 1949). Pan Ku's summary throws no new light on 
it, unless the use of "Liw 7" ("Six Arts") can be proved to 
be a term affected by opponents of the Kung Yang dogmatic 
school. 

We now come to the other two buildings, the Ch^eng 
Ming and Chin Ma chambers, where the writers congre- 
gated. Here Pan Ku's interest is deeply engaged. It is a 
writers' club which he depicts, where, as the Preface informed 
us, they met morning and evening and from day to day dis- 
cussed and took thought. The Ch'ang-an Guest takes us 

[ 96 ] 



PAN KU ON THE WESTERN CAPITAL 

further still in language which surely must represent Pan 
Ku's own mind. These masters of language "went to the 
roots of things." Not only so: they were keenly alive to all 
that their eyes and ears could tell them. Although so many 
of the fu of that time have not survived — yet arguing from 
those which have, and from the titles itemized in the fourth 
of the ju sections in Liu Hsin's Catalogue — there can be no 
question but that the writers' seeing and hearing was in re- 
lation to what they saw with their own eyes about them. The 
description is most impressive and throws an arresting side- 
light on that court which after Emperor Hsuan went down 
hill in such disastrous fashion. Recognizing this, we have 
also to recognize that it was in the days of the decline of 
Athens that Plato and the Academy flourished j the one series 
of events was no bar, indeed was in some ways the very 
stimulating force behind the other series. Although, there- 
fore, we should not idealize the members of this Ch^ang-an 
literary coterie — there undoubtedly were sycophants and 
place-hunters among them — ^yet they must be taken as up- 
holding the banner of free enquiry. They gave themselves 
to exploring the possibilities of language in the interpreting 
and ordering of personal experience. The interesting thing 
is that although they worked apart from the more sober- 
minded scholars and administrators, they at the same time 
had their influence on them, for these latter also "from time 
to time composed," as can be seen from the entries in the fu 
sections of Liu Hsin's Catalogue where the names of various 
statesmen appear as authors. The writers, for their part, al- 
though free from burdensome administrative duties, were 
there on the spot to be used in drafting decrees or preparing 
memorials (cf. Wang Cheung, Lun Heng^ xxiii, 3 and 
xxviii, 2). 

For the rest of the palace area the picture is even more 
in outline. Yet it is instructive as far as it goes. Thus, the 
glorifier of Ch*ang-an can only speak of a medley of build- 
ings where the interior officers of the imperial household 

[ 97 ] 



CRITIQUE 

carried on their duties. Also, the only mention of living quar- 
ters, apart from those of the emperor and the harem, is in 
relation here to the guards. We can infer from other passages 
that officials did find places to sleep when their duties re- 
quired them to be on the spot, but the only possible inference 
is that the arrangements must have been sketchy. Well-sal- 
aried officials had their town houses elsewhere, and the young 
princes, once they were free of their nurses, had their estab- 
lishments outside the palace gates. 

With regard to the officers mentioned as ranking first class 
in the examinations, it was in this period that there began 
the practice of examining candidates for state service. Em- 
peror Wu inaugurated the practice along with the establish- 
ment of his Erudites as a group of teachers officially charged 
with preparing candidates, striplings in the capital of proved 
intelligence and men selected by the heads of the different 
kingdoms, provinces, and commanderies.^ The term Hien 
hstad^^ (the "upright and filial" men) occurs again and again 
in the "Acts of the Emperors" {Pen Chiy in the first cen- 
tury B.C. It was the technical term for suitable candidates, 
and there is no need to infer a very exceptional standard of 
morals in these men. What is important is that there was no 
bar to prevent a man of plebeian origin being recommended. 
As for the nature of the examination, there is good evidence 
that it was in writing and consisted in answering questions 
on the political needs of the moment. The call to the country 
was not made regularly, but came from time to time as need 
emerged for getting fresh blood, or as an emperor thought 
it advisable to restore his credit among the scholars of the 
empire. The phrasing of the passage here conveys the strong 
impression that the people specified were junior officers being 
tried out. Of these the top grade of successful examinees 
went to the Ministry of Rites. For all of them, as the biog- 

^ The development of education is treated in more detail in Ch. XI, 
section 7. 

^ Title of the section in the Han Shu and other dynastic histories devoted 
to court annals. 

[ 98 ] 



PAN KU ON THE WESTERN CAPITAL 

raphies in the "History o£ the Former Han Dynasty" show, 
there was the hope of advancement, and this for the most 
part meant a settled office in some provincial administration. 
As for the "flying galleries" {fei ko)^ they were a private, 
closed-in means of communication with the harem and the 
various buildings mentioned. They must have been elevated 
on struts for them to be called "flying." Doubtless the hug- 
ger-mugger of buildings here described and the coming and 
going of lesser folk made them a great convenience. Gal- 
leries of this kind are mentioned in the "History of the For- 
mer Han Dynasty" (Dubs, I, 113). 

8. The Chien Chang Palace and Its 
Affurtenances 

From the "History of the Former Han Dynasty" and 
from Chang Heng we learn that in 104 b.c. there was a fire 
in the palace grounds, and from another source that in this 
year Emperor Wu began the Chien Chang Palace. Its con- 
struction took a number of years — Pan Ku's description, 
plainly drawn either from detailed records or very observant 
eye-witnesses, demonstrates that it was one of the worst ex- 
travagances of that megalomaniac period. The reader will 
notice the reference to artistic workmanship on the one hand 
and the creation of auspicious influences on the other hand. 
There is the problem we stumbled on in the critique on 
Section 5. In the light of the elaborations planned and put 
into effect in the Chien Chang layout, we cannot but suspect 
that something in the nature of real scientific accuracy was 
necessary, not merely the accuracy of skilled workmanship, 
but even more the accuracy of mind which could plan and 
draw diagrams. We know from a number of earlier books 
that there had been drawing-compasses and T-squares and 
the like in use for at least four hundred years before 104 b.c, 
and the evidence of fine tools for especially delicate work 
is accumulating. Not only so: there is evidence to the effect 
that what the craftsman could achieve by accurate measure- 

[ 99 ] 



CRITIQUE 

ment and calculation had for a considerable time been re- 
garded by scholars as an indication of how the mind should 
be trained to work in the field of ideas (cf. MenciuSy I A, 7). 
That being so, what of the obscurantist aspect of the picture 
as drawn, not merely the construction of centers of auspi- 
cious influence, but also the attempt by pseudo-scientific tech- 
niques to conquer death? 

This last aspect is brought home to us with all the force 
of pungent wit in the description of the terrace, with the 
tower above it and the lakes below. The whole mise-en-scene 
of the Chien Chang site sprang from the fantastic deter- 
mination of one man. Emperor Wu. Spurred on by the magi- 
cians to whom he gave his confidence, he set about building 
this vast conglomeration with precise attention to every preg- 
nant detail, so that in the midst of its luxury he might in 
addition find the way to become an immortal. It was a case 
of viewing Nature as having its laws of cause and effect, and 
of these laws being discernible. Emperor Wu was a man of 
proved powers of statesmanship, able to conceive far-reaching 
plans and entrust their execution to the ablest of deputies. 
This is clear from the "Acts of the Emperors." But alongside 
that story we have to place the revealing story in Pan Ku's 
Chiao Ssu Chih ("Treatise on Sacrifices," Han Shu^ ch. 25 
A and B), where the Emperor is represented as being year 
after year cheated by a set of glib-tongued charlatans. Pan 
Ku's attitude is unmistakeable there, for from time to time 
he makes the caustic comment that no evidence was forth- 
coming. So here in this passage he is completely outspoken 
in his condemnation. On the one hand there was the un- 
conscionable extravagance of the whole proceeding. That he 
makes clear in the first part of his description. In the second 
part in a monotonous succession of six-word couplets he cre- 
ates an atmosphere which is as mentally dizzying as the 
experience he depicts of ascending the terrace and the Ching 
Kan Tower. May we not assume that he was conscious of 
this, deliberately adding one mystic, romantic detail after 

[ 100 ] 



PAN KU ON THE WESTERN CAPITAL 

another in order to excite his readers' amazement — and then 
exploding the fantastic illusion with his "An abode for im- 
mortals — not a place where we mortals can find peace!" Yang 
Hsiung had done much the same thing in his Kan Ch^ilan Fu 
on Emperor Ch^eng's special sacrifice to the T^ai I (Supreme 
One). After a dizzying description of the Kan Chilian Palace 
and the superstitious procedure in the sacrifice, he winds up 
with 

The chosen magicians were all there, yelling at God's 
gate, opening Heaven's audience hall and importuning 
a host of [little] gods: . . . 

... a medley of helpers, bringing down on the altar 
masses of good fortune, heaped up like a mountain. 

Thus it was that with the business finished and the merit 
colossal, we all went back to the carriages and returned 
home . . . 

. . . flying clouds spread in the sky, rain descended every- 
where — and all this was to go on for ten thousand gen- 
erations! 

Here then emerges to view one of the main foci to the 
historian's problem of Han mentality, the mentality which 
was grounded in animism, went on to naturalism, and then 
did — what? In our search for clues we now descry, within 
the circumscribed area of a poet's picture, one side to a really 
great emperor's mentality around lOO b.c. We also descry 
the poet's reaction some hundred and eighty years later. The 
Western reader will, of course, bear in mind what the pages 
of ancient Western history reveal, namely the attraction the 
mystery religions had for educated people in imperial Rome, 
and then, in mediaeval and modern times, the long tale of 
bogus scientists and their dupes in high places. 

9. The Imferial Hunt 

This section with the last and the succeeding one are the 
longest in the poem: surface evidence of the importance 

[ loi ] 



CRITIQUE 

these two aspects of Ch*ang-an life had for Pan Ku, as he 
planned and wrote. The first point to note is his emphasis 
on the political aspect of these periodic hunts. They served 
to keep the troops in training and to warn the barbarian tribes, 
in particular the obstreperous Hsiung Nu in the north and 
west, that the Emperor had a mailed fist. But, having dealt 
with that in a word, the author's pen swings out into the 
description of what he really wanted to bring home. If we 
are tempted when we come to Chang Heng's poems, to re- 
gard Pan Ku as a little superficial, we must do justice to the 
fact that he revolted against the blood bath which he here 
portrays with all the dramatic zest that his Western Capital 
Guest can lend to the account. 

The poetic gift of objectification of the mind, the power 
of putting into the mouth of an imaginary person the words 
which are burning on the poet's lips but which do not ex- 
press his own point of view, was quite clearly not absent 
in the culture of Han times. This fact is of importance in 
the light of what is so well-known among historians of lit- 
erature, namely, that whereas the Greek genius produced 
a Homeric style and the Athenian dramatists, the Chinese 
genius seems to have done neither. This apparent fact about 
China of the classical era needs to be restudied, taking into 
consideration not only these poems but a number of other 
highly trenchant and dramatic poems by the great fu writers 
of the first century b.c. 

Another prevalent notion is that the Chinese people, under 
the inspiration of a Confucius and a Lao Tzu, early evolved 
a highly pacifically-minded culture. There is something to 
be said for this notion, but here is the other side of the pic- 
ture, the side that is made familiar to us at the present day 
by the discovery of the id and other unpleasant phenomena 
in the cultural history of blood sports. There is also the dis- 
covery from certain recent world-shaking events that civi- 
lized man is not so firmly and unalterably civilized as he 
assumes he is. This then is the noteworthy fact here: that 

[ 102 ] 



PAN KU ON THE WESTERN CAPITAL 

huntsmen and soldiers are depicted as unashamedly excited 
by blood-lust, and that the imperial court of the Earlier Han 
is depicted as unashamedly enjoying the sanguinary spectacle. 
The carrying out of the hunt on the scale described cannot 
but have entailed a considerable amount of staff work, that 
is to say, consultation between the forest officials and the 
military, and last but not least the personal secretaries to the 
Emperor. There was an array of beaters, but they would 
have been ineffective unless there were also planners with 
an intimate knowledge of the terrain and the organizing 
power required for concentrating the game at strategic points. 
There is no marked emphasis on this aspect, but two or three 
touches in the account show that it was not absent from Pan 
Ku's mind. It is also well to add that the sport of kings 
and the subsequent revelry was not shared by the peasants, 
and that the strict protection of the imperial forests and the 
savage penalties inflicted for poaching were a direct cause 
of peasant discontent. It is not clear whether the beaters got 
a share of the game, but in any case it would be impossible 
to prevent the professional huntsmen from gossiping to their 
families and friends. It is easy to imagine the effect on them of 
those "heaped-up masses" of dead birds and beasts. 

10. After the Hunt 

The account of the hunt feast rings true to fact. East and 
West would seem to be much alike, and artists will recall 
the Rajput paintings in which rajahs hunt and wassail with 
such verve before our eyes. In the Ch'ang-an picture there 
is a like air of full-blooded jollity about the proceedings, with 
the wine from the wine carts flowing freely. There has been 
no mention of the Emperor or his immediate circle taking 
actual part in the hunt, but we know that princes of the blood 
engaged in the chase, as the Chou feudal lords had done in 
their day, so the tradition was there of a strenuous day in 
the open air followed by heavy eating and drinking. Plainh' 
the stiffness of court etiquette was relaxed for the night and 

[ 103 ] 



CRITIQUE 

the following day. We can imagine the tired huntsmen and 
officers rolling off into a corner to sleep, but for the aristo- 
cratic part of the company the occasion was of the nature 
of a picnic with everyone free to wander at will among the 
"hundred odd'' pavilions, the shady groves and sweet-smell- 
ing plants and flocks of gaily-colored birds. Again the West- 
ern Guest naively gives away the fact that the pleasuance 
in the hunting park had been constructed at very considerable 
expense: a fact emphasized in the Shang Lin Fu (Ssu-ma 
Hsiang-ju). 

The share that the ladies of the Rear Palaces had in the 
picnic is particularly interesting. Not only did they have 
these two days out in the country, with elaborate arrange- 
ments made for their enjoyment, but some of them at any 
rate took their pleasure in quite strenuous fashion by shoot- 
ing with the bow and later climbing to a number of view 
points. These young women were by no means the languor- 
ous houris of Western imagination in relation to oriental 
harems. Also, since appreciation of rugged mountain scenery 
was so late in appearing in the West, here is a very early 
instance of that aesthetic sensitivity. As is well-known, in post- 
Han times it developed continuously, in particular under 
Buddhist influence j here, however, we find it in evidence 
just before Buddhist missionaries arrived in China. 

A third point of interest is the apparent lack of chaperonage 
to these ladies. The narrative has not a word on the subject. 
That, to be sure, affords no ground for assuming that it was 
not there. On the other hand, the whole atmosphere of this 
post-hunt picnic is so impregnated with relaxation and care- 
free gaiety that the historian who knows something of the 
seamy side to the sex life of the Han aristocracy is bound 
to pause here and scrutinize the text and commentaries with 
special care. There is the more need for this since the fu 
writers were master hands at conveying subterranean hints 
on matters of scandal. As far as I can see, there is nothing 

[ 104 ] 



PAN KU ON THE WESTERN CAPITAL 

to suggest that during the night or the day following there 
were any clandestine rendezvous. I take it that there was 
no mention of such in Pan Ku's sources of information, for 
if it were otherwise, it would have suited Pan Ku's book to 
introduce into a fair-seeming narrative a hint of a less repu- 
table side. As the narrative stands, the men played in one 
part of the park, and the women played in another part. 

II. The Western Cafital Guest Sums Uf 

Pan Ku's historiographic ability is here plain to see. The 
details he gives in ordered sequence are masterpieces of clear 
summarization. The picture painted here by this author of 
the "History of the Former Han Dynasty" is that of a great 
state and nation triumphantly solving the problems of gov- 
ernment over a wide continental area, so that life continues 
from generation to generation without economic disaster, po- 
litical disturbances, or invasion from without. The intriguing 
thing is that the section contains no reference to all this pros- 
perity being due to the imperial grace. As the reader who 
studies the poem in the original finds, there is no lack of 
flattering reference to that power of grace in the first of the 
Han emperors. Yet in the final picture not a word! Now an 
argument from silence is rightly regarded as a dubious onej 
but it cannot be avoided altogether, for some silences are 
glaring. Where such a one occurs in a document produced 
by a writer with immense descriptive talent who used a poetic 
genre which entailed unremitting attention to nuances of 
meaning, it is out of the question to pretend that this gap 
of silence is not there. Moreover, part of the success of a fu 
in the Ssu-ma Hsiang-ju and Yang Hsiung tradition would 
depend on the author making his point clear without cross- 
ing his t's and dotting his i's. 

First, what are the unassailable facts? The most outstand- 
ing one is that after Emperor Hsuan's reign (73-49 b.c.) 
this te (mana-w/^j) of the Han regime declined, and 

[ 105 ] 



CRITIQUE 

towards the end declined with portentous speed. Alongside 
of that lies the fact that although the Former Han regime 
had achieved government over an unprecedented extent of 
territory and by its unprecedented machinery of administra- 
tion had brought about an unprecedented prosperity, there 
had been plots and insurrections against the central authority, 
and during and after Emperor Wu's reign impossible bur- 
dens had been laid on the peasant producer. There had also 
been scholar families of repute cut off or brought to poverty 
through court intrigue or owing to malpractice in office. Pan 
Ku knew all this — none better, after his days and nights on 
the palace archives. 

Second, it has to be borne in mind that if the Later Han 
emperors had been lineal descendants of Emperor Wu or 
the emperors after him, it would have been difficult for Pan 
Ku or any other writer of his time to slur their memory. 
For the reigning emperor to tolerate that would be to lay 
himself open to the grave charge of being unfilial. But Kuang 
Wu claimed only to be descended from Emperor Ching, a 
generation above Emperor Wu. That fact would ease the 
situation for a writer who proposed to omit any conventional 
praise of the regime as it became in the last hundred years 
of its tenure. 

Third, the expressed meanings under scrutiny come from 
the lips of the Western Capital Guest, and he has been pre- 
sented in character throughout as a man overwhelmingly im- 
pressed by the material achievements of the regime. The new 
wealth that came into existence, the splendor of the new 
palaces, these are the features of the regime that captured 
his imagination. Pan Ku's intention seems to be clear: it was 
to make the Guest the type of the superficial observer. 

The only conclusion that seems certain is that the omis- 
sion should not be taken as one through sheer inadvertence. 
On the other hand, it surely might well have happened that 
Pan Ku, having arrived at the summing-up stage, decided 

[ io6 ] 



PAN KU ON THE WESTERN CAPITAL 

that there was no great need to go into the unhappy story 
of the final dereliction of the court. Everyone knew it — why 
then talk about it? In any case, it did not fit the text of the 
Guest's sermon. If Pan Ku came to that conclusion, one show- 
ing a sense of dramatic artistry, the only honest and artistic 
course was to omit any reference to the virtus which had been 
so dissipated. As the Guest says in his last sentence, his pic- 
ture "does not give one part in ten." 



[ 107 ] 



Chapter VIII 



CRITIQUE OF CHANG HENG'S 
"FC7 ON THE WESTERN CAPITAL" 

I. The Introduction of Two Interlocutors 

Without a foreword of any sort Chang Heng went straight 
to his self-appointed task. Pan Ku's poems had needed a 
preface, for he was trying out an adaptation of the ju genre 
to a wider and more patently dialectical purpose than any 
of the earlier ju writers had attempted. For Chang Heng, 
with Pan Ku's achievement still in men's minds, the situa- 
tion was different. There was no need to explain any circum- 
stances. The very fact that his composition would bear the 
title Liang Ching Fu (^^Fu on the Two Capitals") would 
demonstrate what he was about j and if any of his readers 
should suspect him of merely imitating his famous prede- 
cessor, the opening sentences, with their characterization of 
two completely new interlocutors, would effectively demon- 
strate the contrary. 

This introductory section is highly dramatic. First, there 
is the kung-tzu^ the nobleman, with the character appellation 
"Stayed-on-Vacuity," and along with him the hsien-sheng^ 
the teacher of a ripe age, by name "Where-did-he-live." These 
two interlocutors are not represented as protagonists, the one 
for a Western and the other for an Eastern capital. Thus 
the center of interest is moved onto a more subtle psychologi- 
cal level than in Pan Ku's poem, a level on which differences 
of temperament and moral outlook create the dialectic of 
drama. For the moment Chang Heng is content to give a 
thumb-nail sketch of the nobleman j it is only in the ^^Fu on 
the Eastern Capital" that the teacher becomes more than 
a dim lay figure. There he becomes a person, in many ways 

[ 108 ] 



CHANG HENG ON THE WESTERN CAPITAL 

representing Chang Heng himself in a fashion that the Hon- 
orable Stayed-on-Vacuity does not. Yet who can venture to 
affirm that Chang Heng did not recognize in himself cer- 
tain strains which were on a par with the theorizing of his 
nobleman? The very fact that his "teacher" is characterized 
as an unlocatable person, as one who may be more up in the 
air or more with his feet on solid earth, points in this direc- 
tion. Living when Chang Heng did, under the emperors he 
served, he would have been a poor poet had it been otherwise. 

Second, the nobleman opens the proceedings by laying 
down the law in arrogant fashion to the teacher, presumably 
some years older than himself. He sets forth a logical argu- 
ment by which he proves a proposition which we have good 
reason for believing Chang Heng profoundly doubted. Thus 
the argument requires the closest scrutiny, for it represents 
the poet's power of dramatic imagination. He could put him- 
self inside the nobleman's type of mind and argue solemnly 
from that point of view. 

Third, the argument is based not merely on naturalistic, 
but also on materialistic pre-suppositions. The nobleman's 
"Heaven" is not the transcendent Heaven with an operative 
will for man's good as characterized by traditional religion. 
The "Heaven" and "Earth" here are the mechanical cause- 
and-effect forces of the Yin- Yang cosmology. Also, the key 
to the argument is the soil, in the one case more productive, 
in the other case less so. In other words, the universe is 
assumed to be a non-moral one in which the economic factor 
is decisive. That was perfectly recognizable to Chang Heng's 
readers, since it was the stark. Legalist position, and that 
theory had its influence on some of the Ju. 

Fourth, the argument proceeds on the basis of a strict 
analogy between small and great nations: an assumption 
which follows from recognition of Nature's mechanical proc- 
esses. The dialectic would easily carry Han readers in assent, 
as it proceeds to "/i che^^ (the class of persons called "em- 

[ 109 ] 



CRITIQUE 

perors"), and to the propositions given under that heading 
and that of ^^chao mm^^ (the mass of the people). 

Fifth, the final stage is an appeal to history in which the 
fall of the Chou order is accounted for entirely in terms of 
economic resources. That from a Ju standpoint was only half 
the truth of the matter. Are we not compelled, therefore, 
to assume that the parallel assertion about Former Han and 
Later Han were for Chang Heng equally specious? As we 
probe into the underlying cynicism of the statement, this 
becomes doubly clear. "Emperor Kao had his capital in the 
west and was lavish in his expenditure. Emperor Kuang-wu. 
. . ." In other words, the one regime had the money and 
so could use itj the other regime had not got the money 
and so could not. Just that, and no more, with everybody 
knowing what misery had come about from the Government's 
prodigal expenditure! 

It is assumed, therefore, that here at the outset Chang 
Heng gives a study in sophistry, the sophistry of a mind 
which is essentially Legalist. In confirmation of this assump- 
tion it is to be noted that the whole passage is in "scripture 
style," the aphoristic, didactic style so characteristic of the 
dialectic in the Scriptures, e.g. in the later "Wings" of the 
"Changes" Scripture. The use of the style here would seem 
to be deliberate, since the rest of the ]u is predominantly 
rhymed. As will appear later again and again, Chang Heng 
had a genius for irony. So here his poet's pen is barbed. 

2. The Tofografhy and Geografhy of Ch^ang-an 

The story begins with the topography of Ch*ang-an, just 
as Pan Ku's does. Chang Heng's description follows the 
same orderly method, taking the four points of the compass 
in turn. The points in which they differ as far as I can see 
are not important j but that is a matter for the geographers 
to settle. Our point of interest is that both authors, the his- 
torian and the scientist, had a trained eye for typographical 
features. 

[ no ] 



CHANG HENG ON THE WESTERN CAPITAL 

With regard to Chang Heng's mentality here, he redeems 
the dullness of his topographical narrative by the lively story 
of Duke Mu of Ch4n (638-619 b.c). It is already clear 
that Pan Ku could let his pen run on a good story, and 
indeed a good fu in those days was expected to amuse as 
well as instruct. This side of Chang Heng's nature comes 
out forcibly enough in both his /^, so we may well doubt 
here whether our learned scientist took the story he told to 
be historic truth. The moral of the supernatural gift to the 
duke, it should be noted, is in full keeping with the noble- 
man's theory of economic causes as normally determinative. 
When, however, the narrative reaches Emperor Kao and his 
abandonment of Loyang in favor of Ch^ang-an, a super- 
natural factor is introduced, namely. Heaven's direct inspira- 
tion of the unknown exile from frontier duty, Lou Ching. 
In the "History of the Former Han Dynasty" account of 
this (ch. 43) there is no mention of heaven-given inspira- 
tion, only stressing of the curious means by which Lou Ching 
managed to see the Emperor and the lightning effect his 
counsel had. It appears, then, that the stress on inspiration 
is Chang Heng's own, and we are faced with the question 
whether he really believed this. Not necessarily so, for it 
is the nobleman speaking j but all our knowledge of ordinary 
mid-Han educated mentality leads to the conclusion that 
emperors were regarded as specially subject to miraculous 
direction of one sort and another. 

3. The Palace Buildings 

Instead of describing next the city and its throbbing life, 
as the prototype did, Chang Heng, after just mentioning 
the city walls, went straight to the palace hill and its build- 
ings, relegating description of the city to be the next center 
of interest. Both authors would seem to have had good ar- 
tistic reasons for their respective arrangements, and in Chang 
Heng's case the narrative goes smoothly from remarks on 

[ III ] 



CRITIQUE 

Kao Tsu's purpose to what he and his successors did in the 
way of glorifying their capital. 

There are forty-nine couplets devoted to description of 
the buildings on the palace hill : twenty to the main residence, 
the Vermilion Palace with its different ramifications, four- 
teen to a miscellany of other buildings, and eleven to the 
women's apartments. Chang Heng's account makes it clearer 
that Emperor Kao's Wei Yang Palace was added to later. 
His description of its extravagant ornamentation is like Pan 
Ku's in its piling up of the adjectives, although here again 
he gets a slightly clearer picture across. His concluding state- 
ment dealing with the Audience Hall at the east end and 
the Jade Tower and K^un Te building at the west end seem 
to have a sting in the tail: "with their irregular elevations 
. . . , any pattern of design was unrecognizable." 

The account of the other buildings is not much help, any 
more than was Pan Ku's description: plenty of names, and 
they of an animistic-poetic nature, but no indication as to the 
layout of the departments of state. Hsieh Tsung (d. 243) 
and the T^ang commentators have no information on this 
score. However, the information about the Emperor's per- 
sonal attendants and the departmental officers being on duty 
at night in shifts is worth having. With this elaborate bu- 
reaucracy at work and couriers coming and going, the palace 
hill must have been a scene of bustling activity, both through- 
out the day and sometimes in the night. Chang Heng lifts 
a corner of the curtain, but is a little puzzling, for he makes 
no mention of Pan Ku's Ch^eng Ming building, but does 
refer to a Lan T'ai building which Pan Ku omits. Since we 
know from other sources that the Lan T'ai was used as 
another storing place for records, the notable feature of Chang 
Heng's account is that it agrees with Pan Ku's in regard to 
the places for the collating of records, but has nothing to 
say on the writers meeting and discussing in the Ch'eng Ming 
building. That seems very odd until we remember that those 
writers were salaried officers in some sort of secretarial posi- 

[ 112 ] 



CHANG HENG ON THE WESTERN CAPITAL 

tion. In those circumstances they presumably took turns on 
night duty, and we can safely assume that the conversation 
was by no means all on routine business. The only question 
is why Chang Heng left this side out. Possibly he felt Pan 
Ku had overstressed this feature. 

4. The Rear Afartments 

Here, as in Pan Ku's poetic effort, there is a luscious de- 
scription of the harem. Whether both poets regarded these 
luxuries as rather barbaric will be considered later. It is an- 
other aspect which requires stressing: the emphasis laid on 
the craftsmen and their triumphs. As in imperial Rome so 
in Chang Heng's Ch^ang-an, the concentration of unprece- 
dented wealth in one man's hands gave an immense stimulus 
to the decorative arts and unquestionably brought new refine- 
ments of aesthetic appreciation. 

As for the frivolous remark about looking inside — suitable 
enough coming from an aristocrat — to the student acquainted 
with the ritual books of that age it may well cause surprise. 
Indeed, one would have thought that such a suggestion might 
easily cause imperial displeasure. Imagine that dour prude 
Empress Dowager Teng having her attention drawn to this 
offence against propriety. As a matter of fact, things might 
be said in relation to a previous regime which might not be 
said about the present. Also, Han literature is by no means 
lacking in evidence of male susceptibility to feminine charm. 
Chang Heng himself is an instance. In his ju on thought 
transcending the normal he uses a delightfully light touch 
in describing how two goddesses tried to ensnare him, and 
what a beguiling song they sang. 

Then comes something outside Pan Ku's sphere of inter- 
est: the psychological effect of these various sense enticements 
on the emperors themselves. In five tense four-word couplets 
the picture is painted with trenchant strokes: a constant round 
of indulgence of one sort and another, every day some new 
beauty the Emperor had not seen before. There is of course 

[ 113 ] 



CRITIQUE 

a flourish of poetic hyperbole here. The wheels of govern- 
ment had to be kept moving, even though His Majesty might 
forget to go back to his own apartments. We are reminded 
of two people specially in Chang Heng's mind, Emperors 
Wu and Ch^eng. The former, it is on record, spent contin- 
uous nights and days in his harem, and because of this made 
the innovation of having eunuchs as private secretaries — 
much to the disgust of the bureaucracy. 

5. The Outbreak of Megalomania 

Again the psychological interest is to the fore, with this 
subtle transference from sensuality to the megalomaniac urge. 
The lurking irony in the last section becomes more visible 
right from the start. That "divine beauty" (shen U) gives 
the key. The T^ang commentators give no reference to Hsun 
Tzu, but there is a striking resemblance to that book's theory 
that the Son of Heaven should be surrounded at all times 
by the most aesthetic sights and sounds. So will his te (mana- 
spirituality) be cultivated to its highest pitch. We see the 
idea at work here, in very realistic fashion, from the point 
of view of a Later Han satirist. Again it is Emperor Wu in 
the first place who is in his mind, although some of the later 
emperors were tarred with much the same brush. Emperor 
Wu not only built with reckless expenditure j he also com- 
pared himself to the Ch4n First Emperor. That monarch 
was the mightiest phenomenon in real history. Hence it be- 
hooved his Han compeer to demonstrate that his shen U was 
on a grander as also a more refined level. Was not his empire 
within the four seas much wider? As a side-light on the 
Former Han attitude to the Ch4n regime, and, yet more, 
as a side-light on the emperor who brought current Confu- 
cianism to the fore, here is a useful directive for the historian. 

The Kan Chilian Palaces, also initiated by Emperor Wu, 
are merely touched on. Chang Heng's readers would be 
familiar with Yang Hsiung's famous description in his ironi- 
cal Kan Ch^uan Fu, where he described the effect on him 

[ 114 ] 



CHANG HENG ON THE WESTERN CAPITAL 

of the "great palace buildings, a fantasy of clouds and billows 
spread to view . . . my eyes fluttered back and forth, forth 
and back, my two souls made small and brought to stupe- 
faction." This was a reference to Emperor Ch^eng's time, 
when several emperors had added their respective contribu- 
tions. Kan Chilian with its amenities became a rival to Ch^ang- 
an, and thereby created a number of problems of secular 
and religious administration. According to Yang Hsiung it 
had a site with extraordinary demon-defeating powers. In 
this the great tower above the valley played its partj so novel 
and striking a feature as it was in the landscape, and so reli- 
giously reassuring with its cleaving of the sky. There is no 
doubt of Chang Heng's appreciation of the constructional 
problem it entailed. He emphasized two points, one that its 
planning went beyond "all ordinary computation," the other 
that it was tapered with precision. 

6. The Chien Chang Palace Area, with Its 
Sfirit-Revealing Tower 

The plot thickens as Chang Heng in his ironical vein takes 
us to the constructional operations outside the western city 
wall. Pan Ku had done full justice to this theme, both in his 
general account and in his witty description of a man fight- 
ing giddiness as he stumbled up and down those interminable 
stairs. Chang Heng does not compete with his prototype 
there. As the man who invented a seismograph, Chang Heng 
appreciated the problems which Emperor Wu's master-builder 
had to face. His description is a masterpiece in its succinct- 
ness and inclusion of essential detail. 

What we were only able to suspect in the critique of Pan 
Ku's ^^Fu on the Western Capital," Section 8, takes shape 
now as more than a mere conjecture. As we put the language 
here alongside his statement that the Heaven-Communicating 
Tower (at Kan Chilian) "surpassed all ordinary computa- 
tion," the demand on the master-builder's ingenuity in experi- 
ment becomes clear to view. The two towers were built at 

[ 115 ] 



CRITIQUE 

the same time, the Kan Chilian one 30 chang high, the Chien 
Chang one 50 chang (in terms of English measurement about 
225 feet and 375 feet respectively), the one a solid structure, 
the other built on the ching kan ("well-head") principle. 
That is to say, the master-builder decided from experience 
with the Heaven-Communicating Tower that to go up an- 
other 20 chang with a solid structure was out of the question. 
The only way was to have an erection of wood and that after 
the pattern of the protective criss-cross wooden structures 
used in relation to well-heads. The danger to a solid struc- 
ture from a high wind was too great : as Yang Hsiung wrote, 
"a gust of wind with startling force . . . leveled the boughs 
of the willows." Therefore an open-work structure was neces- 
sary, either four- or eight-sided. Then came the problem of 
weight of timber. Heavy pillars would make the upper stories 
too heavy, the whole thing would come crashing down. 
Therefore light pillars must be used. On the other hand, 
as the Greek and Roman builders discovered with the sup- 
port needed for heavy superstructures, the ninety-degree 
angle of a pillar did not afford the necessary rigidity of 
support. In Ch^ang-an this problem was solved by using 
pairs of brackets^ clamped to a regular open succession of 
beams and their supporting pillars. Thus the principle of 
counter-acting pressures was brought into action. There is 
still the question whether an open-work, light-timbered struc- 
ture could have achieved the amazing height of a hundred 
stories. The problem is one for engineers to decide. The in- 
teresting thing is that the Pharos of Alexandria, built by 
Sostratus of Cnidus in the third century b.c, was believed 
in classical times to be six hundred feet in height. Whether 
that was entirely a solid structure is, I gather, doubtful, but 
in any case it was built to last. Emperor Wu was not so 
much concerned about that. 

Masters of the works and master-builders are, unfortu- 

^ Wood bent and stiffened by being soaked in water, a practice employed 
for making certain parts of the crossbow. 

[ 116 ] 



CHANG HENG ON THE WESTERN CAPITAL 

nately, seldom mentioned by name in the literature of the 
time, nor are the artist craftsmen. Section 4, however, has 
the intriguing statement (omitted in the translation) : "or- 
ders were given to [? a] Pan and [? an] Erh." This was 
in relation to the construction of the flying galleries (and 
possibly the Rear Palaces). By Han times one Lu Pan seems 
to have been well on the way to becoming the patron saint 
of workers in wood. One legend has it that he was of princely 
descent and lived at the end of the sixth century b.c, being 
master-craftsman to the Duke of Lu. There were also a num- 
ber of stories about him as the inventor of certain wonder- 
exciting mechanical toys. Wang Erh is more of a mystery 
figure, but is referred to in Huai Nan Tzu (second century 
B.C.) as a master in the use (? the inventor) of a curved 
wood-carving knife and a rounded gouge or chisel. Yang 
Hsiung in his Kan Ch^ilan Fu refers to one Ho as the author 
of the dumbfounding patterns of rococo decorating there, 
and adds that "if Lu Pan and Wang Erh were to see them 
they would throw away their tools." It is evident that Yang 
Hsiung thought very great advances were being made in 
the arts and crafts. Couple this with Chang Heng's descrip- 
tion of those timbers and the clamps and the invention of 
a revolving weather vane and draught-producing shutters, 
and we seem to have quite a little reliable information about 
mechanical developments in the Han dynasty. Also the Chung 
Yung ("The Mean in Action," ch. 20) comes to life in this 
connection: "If daily and monthly trials of skill be held, 
and grants of better rations be given on the merit of the 
work done, the hundred crafts are thereby encouraged." 
The "Shaman from Yueh" ("Y^VA wu^"^) (or, as Dubs 
maintains, "shamaness") should catch the historian's eye. The 
great Emperor, so ruthless in his plans to absorb the sur- 
rounding peoples, could nonetheless be led by the nose by 
their barbarian beliefs. Chang Heng as a Later Han man 
could feel that, and there can be little doubt but that his 
reference here was meant to have a sarcastic edge. Whether 

[ 117 ] 



CRITIQUE 

the founding Han emperors had anything like the same 
sense of superiority is doubtful, and it still remains to be 
seen whether further critical studies will not reveal that bar- 
barian beliefs and barbarian mythologies penetrated a good 
deal further into the Sinitic culture than later ages realized. 
As for barbarians at court generally in Emperor Wu's time, 
the words "Tu Lu" reveal the presence of members of that 
tribe in Ch'ang-an. The Tu Lu were a Turkish people oc- 
cupying territory a thousand miles west of Ch^ang-an, and 
acrobats from there were a feature in Han court entertain- 
ments. One of their stunts was to climb a pole and then fling 
themselves higher and catch the pole as they fell.^ Thus 
*; Chang Heng's poet mind had an apt image to hand to clothe 

p his ironical intent. 

\\ The final reaction to Emperor Wu's carefully mobilized 

*} attack on the problem of transcending the mortal sphere is 

\\ half one of sympathy. The poet knew by his own experience 

Cj what it was to long for the Beyond. In his Ssu Hsuan Fu 

he described how eventually he was transported beyond the 
sun and stars and the Milky Way, until he looked out on 
chaos, on elemental material forces whirling and colliding 
in the Ultimate Beyond. Thus the tenor of his "before you 
have gone half-way, you are shaking with fear . . ." is exactly 
en raffort with the tenor of that poem. The Honorable 
Stayed-on-Vacuity is portrayed as jeering at Emperor Wu 
for believing that commensurable, mechanically-acting means 
could be used for prolonging his natural span in life. Yet in 
the opening sentences of the poem the determination of man's 
lot by material, mechanically-acting forces was the very theory 
that the nobleman proved — at any rate to his own satisfaction. 
Perhaps the main point of interest here lies in the amaze- 
ment, expressed with uncompromising bluntness, that an 
educated, intelligent person could be so led astray by the 
obvious absurdities handed out by charlatans like Shao Chiin 

2 See Han S/iu, ch. 96 (Hsi Yil Lieh Chuan) and the Cheng Tu Fu of 
Fu Hsiian (a.d. 217-278), ap. / Wen Lei Chii (ch. 61). 

[ 118 ] 



i? 



CHANG HENG ON THE WESTERN CAPITAL 

and Liian Ta.^ We have here the evidence of a Later Han 
rational reaction against the more glaring superstitions of 
the earlier age. The historian's question is whether that age 
was definitely more sunk in superstition. It appears that it 
was. Apart from Emperor Wu's aberrations from common 
sense — and he, it must be remembered, was a fully literate 
man of great executive ability — the high spot of naturalistic 
speculation in his age was Tung Chung-Shu's semi-rational 
Yin-Yang theories, and those theories not only captivated 
the scholar intellect during the succeeding century but also 
had a potent influence in the first century a.d. The Kung Yang 
Chuan^ with its intense preoccupation with prodigies and 
portents, had overwhelming support in the White Tiger 
Conference in a.d. 79, as it had previously had in the Shih 
Ch^u Conference in 53-51 b.c.j and the production of the 
apocryphal "Woof Books" went on all the time. Yet the 
Later Han era produced not only a Chang Heng but also 
that notable iconoclast, Wang Cheung, whose Lun Heng pil- 
loried the current affronts to commonsense and reason. Add 
to that Chang Heng's expectation that his court readers would 
appreciate his slashing sarcasm. There was a new leaven at 
work, traceable in the works of Yang Hsiung (53 b.c.-a.d. 18) 
and Liu Hsin (d. a.d. 23)3 and by the end of that century 
the new wine poured into old bottles was bursting the more 
decrepit ones. So much by way of an interim decision on 
this intricate problem. 

7. The Chien Chang Area (continued) 

The two accounts confirm each other in outline. Chang 
Heng reveals again his powers of orderly observation, the 
evidence here for this being very striking. The narrative pro- 
ceeds as if it were from a bird's eye view: first the two towers 
with their geomantic layout, then the great huddle of build- 

^ Pan Ku satirizes them in his Chiao Ssu Chth ("Treatise on Sacrifices," 
Han Shu^ ch. 25). Shao Chiin had an astounding-ly long- run for his money 
and died in good favour with his master. Liian Ta, an even more spectacular 
success, was caught in the end and executed. 

[ "9 ] 



CRITIQUE 

ings which constituted the main palace, then the interlacing 
paths in all directions and the covered passage running east 
to the city wall. Then in the distance come the lakes and 
their fairy hills j and so we reach the hills of the hunting 
park on the horizon. Poetic power of imagination is there, 
but that genius so much more often than some theories allow 
is rooted and grounded in precise information, impressed, 
it may be, on the poet in his early youth. Thus Chang Heng, 
born in a.d. 78, was not so far removed from the Ch^ang-an 
period that he could not have heard from some old man, 
e.g. his grandfather, his impressions of Chien Chang before 
it went to rack and ruin. In any case the picture is vividly 
before the poet's eye: "as you gaze you cannot find the way 
back." This phrase should be compared with Pan Ku's "gaze 
out of focus." Both remarks entail a cultivated level of con- 
sciousness in relation to the optic sense, a particular lucidity 
of mind for which the two men's linguistic medium afforded 
an adequate channel of communication. 

8. The City and Its Suburbs with 
Their Inhabitants 

May we assume that Pan Ku's vivid picture of the shop- 
ping center throng pleased him and his readers as much as 
it pleases us? It was the first time there had been such a city 
with so many stores and buyers. He exaggerated, more or 
less, and since one poet's hyperbole may easily leave another 
poet cold, Chang Heng got the same effect home by other 
means. He first drew attention to the occupants of those 
magnificent houses in the exclusive quarter at the north end. 
Shih Hsien and Tung Hsien were the outstanding examples 
of the parvenu in the earlier court, and while they lasted, 
they had money to spend in the luxury stores. (No mention 
is made of other families there, but they must have been 
few, for most of the nobles and great officials had their houses 
by the imperial tombs.) The account then proceeds to the 
rich merchants produced by the trade boom and, as it pro- 

[ 120 ] 



CHANG HENG ON THE WESTERN CAPITAL 

ceeds, we learn in a sentence that everything the heart could 
wish was there on sale — and profits ran to a hundred per 
cent. Another sentence gives us the hawkers, and the cheat- 
ing of "small-town folk." 

With regard to the parvenus, Shih Hsien started his career 
as an office-trained junior, fell foul of the law and was cas- 
trated, attracted the attention of that simple-minded monarch. 
Emperor Yiian, and so became his eunuch in the Rear Palace. 
His influence was unbounded, and extremely injurious to the 
state. Tung Hsien was not a eunuch, but a handsome young 
man with whom Emperor Ai fell in love. As to the extrava- 
gant houses these two men built — one of them was known 
by the same name as an imperial palace — Chang Heng's lan- 
guage is very reminiscent of what Pan Ku's story had re- 
corded, so that we know the source of his information. The 
stands of arms at the entrance were of course a necessary 
precaution. Shih Hsien incurred enmity in many quarters 
by compassing the death of men whom he feared j and, as 
the story here shows, there were plenty of gangsters about 
ready to be hired. 

With regard to the wealthy merchants, considerable detail 
on this class is to be found in the Shih Huo Chih monograph 
in the "History of the Former Han Dynasty" and Dr. Nancy 
Lee Swann's careful translation and annotation of it. Chang 
Heng names four members of this class: one who made his 
money in fats, one in selling cooked tripe, one in opening 
up iron mines in Szechuan, and the fourth by horse doctoring. 
Since there were valuable horses from Ferghana in the im- 
perial stables and those of the aristocracy, a skilful veterinary 
would be in continual demand. The odd thing is that luxury 
goods do not appear in this list, whilst a vendor of cooked 
tripe does. Presumably there is some satirical meaning be- 
hind this. The sense of class distinction between the court 
and the city was a marked one, and a relatively poor scholar 
gentleman like Chang Heng would hardly be immune from 
it, the less so because many of the forty dealers we know 

[ 121 ] 



CRITIQUE 

of were operators in the money market in direct contraven- 
tion of the law. 

Where Pan Ku makes a brief and colorless reference to 
the desperadoes, congregated in the capital from all parts 
of the country, Chang Heng goes into significant detail. He 
gives the names of four men who not only "formed gangs" 
but also "joined forces and united their intentions without 
jealousy." His summing up of them is dramatic enough: 
"spite-fire eyes over a trifling affront — a dead body in an 
angle of the road." Add to that the sobriquet "Yang of the 
Mou Tombs and Chu of the Yang Tombs, quick to quarrel, 
roaring in a flash." Such is the glimpse we get of the gangster 
underworld of Ch'ang-an. 

Chang Heng's lighter, ironical vein reappears in the treat- 
ment of the argumentative, out-of-work scholars. It reminds 
one of Plato's criticism of the sophists, as also of Helen 
WaddelPs description of the students in Abelard's Paris. We 
know that in the century after Emperor Wu's inauguration 
of empire-wide recommendations and tests in the capital for 
government service, the numbers of candidates went up by 
leaps and bounds, the total running finally to tens of thou- 
sands. Of these Chang Heng refers here only to those of the 
five home counties — whether prospective candidates or suc- 
cessful ones is not clear. The latter, although given the 
status of langj did not necessarily get posts right away. They 
went into a waiting list, and they might have to wait a very 
long time. Men in that situation doubtless would congregate 
at the capital, so as to be able to approach what influential 
persons they knew. Meanwhile they argued at street corners. 

That this was a highly argumentative age in the scholarly 
world has already become clear in the last chapter. But what 
actual topics caused the fury of discussion depicted here is 
not clear at all. Chang Heng's "splitting of hairs and tearing 
of muscles apart," illuminating as it is as to the temper of 
the rank and file, and his slightly contemptuous attitude to 
that temper, helps us only very little. In view of the contents 

[ 122 ] 



CHANG HENG ON THE WESTERN CAPITAL 

of the three chapters on the famous scholars in the two Han 
Histories, coupled with the contents of the Liu Hsin cata- 
logue, the center of interest was the various texts of the 
Scriptures, each with its own ardent band of supporters. That 
being so, the splitting of hairs and tearing apart of muscles 
presumably was over the variant readings in those texts. 
Questions of single words and the proper punctuation were 
enough to produce violent disagreement. 

Taking these three phenomena together, merchant mil- 
lionaires, criminals and gangsters, and out-of-work scholars, 
we get a side-light on the social order of the time. It is not 
enough to realize that however rich a business man became, 
he could not buy his way into the ranks of the nobility. 
There was, however, a way open to him to raise the social 
status of his family. He could have his sons trained for 
government service, and his commercial dealings with aristo- 
crats and officials would give him the opportunity for urging 
their recommendation for posts. Once a man was started on 
an official career there was no bar to his reaching the status 
of enfiefed nobility. So with regard to the underworld, these 
bands of lawless desperadoes and swashbucklers, some of 
whom took service in noble families: where did they and 
their skill in arms come from? When we remember that 
training in the use of arms was part of the education of a 
nobleman, that plotting for power by the royal princes and 
their friends again and again brought not only death to the 
ringleaders but reduction of their descendants to plebeian rank, 
we begin to see one source from which the desperado class 
was fed. And how about the princely families that the First 
Emperor demobilized and whose memory of happier days 
may be connected with the Han surname Kung-sheng} Did 
some of their descendants cherish memories of their aristo- 
cratic past and inherit some of their ancestors' arms? And 
how about the descendants of the imperial Lu clan who after 
five generations had no claim to kinship with the reigning 
emperor? So with the tens of thousands of candidates for 

[ 123 ] 



CRITIQUE 

office in the last years of Former Hanj for those with the 
ability and luck to get and hold appointments, they and 
their children would go up in the social scale. The rest, if 
from official families, might easily go down. There is, therefore, 
no substantial ground for taking such pronouncements as those 
in the Wang Chih of the "Record of Rites" (dateable for 
the main part of it as the product of Emperor Wen's reign, 
179-157 B.C.) or those of the Chou Li (less easily dateable), 
as representing a fixed order of society, the members of which 
could not move up or down in the social and economic scale. 

9. The Hunting Park 

In Pan Ku's poem there is no description of the hunting 
park itself, no itemization of what was in it. Chang Heng, 
on the contrary, gave his readers an elaborate picture. First 
the main landmarks by which they could estimate the huge 
size of the park, then a very detailed account of the contents, 
the flora and fauna in their natural setting, with a glimpse 
of the surrounding wall and the monstrous number of pleas- 
ure buildings. He noted the migration of birds in due season 
and wound up with the care the foresters used in attending 
to the welfare of the animals in winter. This painter's eye 
for landscape had appeared in both Ssu-ma Hsiang-ju's and 
Yang Hsiung's hunting fw. Ssu-ma Hsiang-ju splashing on 
his colors enthusiastically j Yang Hsiung equally intoxicated 
with language but careful to give a historical setting. Chang 
Heng's picture in comparison is more methodical than theirs. 
What was it in that mid-Han period which aroused this pas- 
sion for detailed description, for what Lu Chi in a.d. 302 was 
to call "/^i TO«" — "embodying an object of observation"? The 
same epoch produced the first draft of what became the Pen 
Ts'ao Ching ("Classic of Fundamental Herbs") with its 2^s 
plants, as also the Shan Hai Ching ("Classic of Mountains 
and Seas") with its thirteen main divisions in which were 
plotted out the mountains and watercourses of the empire. 
There is also the ^'Tribute of Yil^^ — was that too the fruit 

[ 124 ] 



CHANG HENG ON THE WESTERN CAPITAL 

of this new talent of the mind, the talent which in Rome 
first emerged in the writings of Julius Caesar and Virgil? 
This section is more than a plain picture: it has a strong 
satirical vein running through it, starting from the opening 
words. To speak of a hunting park as an enclosed, interdicted 
chi was a contradiction in terms, a contradiction reinforced 
by the second clause in the couplet, "under the control of 
the Intendant of the Capital," i.e. not of the Son of Heaven 
himself. The words which rang in Chang Heng's and his 
readers' heads were ^^fang chi chHen //, wei mm so chlh^'^ "the 
national demesne of a thousand U [300 odd miles] is where 
the people are at rest." So went the ringing lines in the Shang 
sacrificial ode ("Odes," Shang Sung). What is even more 
significant here, the tale was told of Confucius that in teach- 
ing his disciples he enlarged on these lines with reflections 
on the huang bird twittering happily on its resting place, and 
then asked whether man should be less intelligent than a bird. 
So we can almost see our poet-satirist with brush in hand, 
cogitating on the Ch^ang-an chi which was not a chu He 
would recall Yang Hsiung's Yil Lieh Fu Foreword which 
pillories Emperor Wu for enlarging the interdicted enclos- 
ure, and his pungent comparison of that act with the Sage- 
Kings whose rule "did not rob the people of rich crop land, 
of mulberry and ch^e trees [so that] women had ample cloth 
and men had ample grain." Then down came the brush on 
the tablet, and he wrote ^'jeng ch'P not '^fang cht^'^ and in 
the place of a people at rest under the Emperor's beneficent 
care gave the image of the controller of the park, the man 
whose business was to punish ruthlessly infringements of the 
interdict. 

10. The Imferial Hunt 

Considerations of space forbid a detailed comparison of 
the two accounts, and all the more any consideration of the 
dependence of our two authors on Ssu-ma Hsiang-ju and 
Yang Hsiung. In general, Pan Ku's story gives a better sense 
of the organized stages of the hunt. At the same time the 

[ 125 ] 



CRITIQUE 

old timer of Ch^ang-an rather revels in the details of carnage 
and destruction. That is not to say that Pan Ku's intention 
was merely to paint the sensational picture in all its gory 
detail. That would be wrong, for in the opening section of 
his "F^ on the Eastern Capital" the Loyang host jeers at 
his guest for being a true son of barbarian Ch^n. Turning 
to Chang Heng's nobleman, he also is in character, but as 
his story proceeds, paragraph by paragraph, the Later Han 
age's contempt for the earlier ethos is more openly displayed. 
Also the nobleman's creator comes right out into the open 
as a man of feeling. 

We now see why in Section 9 he described so carefully 
the birds and beasts in their comfortable habitat j the time 
was at hand when these same birds and beasts would be flee- 
ing in blind panic and be overtaken by an untimely death. 
And if a stray one should be lucky and escape to the suburbs, 
it would fall a prey to a falconer or a dog on a leash. We 
see here, then, not only the literary artist in command of 
the vivid phrase for depicting the scene in all its picturesque 
and disgusting phases, but also the humanitarian spirit, quick 
in its appreciation of the torture that goes with panic, subtle 
in its delineation of ferocious beasts brought to bay. 

It is poetry of a high order here. In the first place the 
sentence structure ranges jaggedly from four-word couplets 
to seven, from seven to three, and vice versa, giving the feel 
of the hubbub and confusion. In the second place, the narra- 
tive progresses from point to point with certain subtle con- 
trasts of theme: from the work on Yii Ch^u's magic tech- 
niques to the imperial guards on parade, from the poor harm- 
less pheasant and hare to the terrifying ssu (? rhinoceros) 
and tigers, and from them to the berserk feats of the pro- 
fessional huntsmen. In the third place, there is the unfor- 
gettable imagery at key points: "the stiff corpses of the slain, 
bird and beast, glistening like tide-washed pebbles" j "before 
the sun is down and [the shadows] change, the killing is 
done — out of ten, seven, or eight." 

[ 126 ] 



CHANG HENG ON THE WESTERN CAPITAL 

With regard to the books of magic and the "lively tales," 
the passage is highly revealing in more ways than one. There 
is, for example, what Li Chou-han noted in his comment as 
"covert opposition" to these books. That is putting it rather 
mildly. From the first word "It was not only play" through 
"what was wanted in relaxation" down to Ch^h Yu, the 
resuscitated magician of legend and the demons, not one of 
which could now cross the Emperor's path, the scorn is com- 
pletely overt. The picture here is almost too compact, oscil- 
lating as it does between the serious business of preparing 
charms and the lively tales for idle moments. However, with 
the help of Liu Hsin's Catalogue the passage becomes per- 
fectly clear. The catalogue lists fifteen collections of "lively 
tales," the last and by far the largest of which is "Y^' Ch^u 
Chou Shuo in 943 sections,"* and Liu Hsin's comment con- 
tains these words: "market-place gossip, alley-way talk, what 
is concocted by listeners on the roads and talkers on the 
streets." 

As for Ch4h Yu with his axe and shaggy coverlet, clearly 
it was through his agency that the imperial party came to 
know the "malignant tricks of the spirits" {shen Men) and 
could take due precautions. 

II. A]ter the Hunt 

It may well be that this enormous section (as I make it) 
should be divided into three or more sections. There are, 
however, prosodic difficulties to doing so. Also Chang Heng's 
intention is clear, namely to pile Pelion on Ossa with regard 
to these festivities, and so be able the more effectively to play 
the critic. In this respect his account strikes a challenging 
note which is not found in Pan Ku's version, and the dra- 
matic interest is intensified with greater subtlety. Thus, for 
instance, the center of interest shifts from the general to the 
particular: first the ladies of the harem, then the rank and 
file at their feasting, the ordering of the carriages, the court 

* See Wang Hsien-ch'ien, Han Shu Pu-chu^ ch. 30. 

[ 127 ] 



CRITIQUE 

circles and their shooting and fishing. Finally the reader's 
attention is concentrated on the Emperor himself, in his 
luxurious lodge by the arena, as a spectator of the grand 
entertainment — and as a human being surfeited with pleasure 
and riddled with ennui. 

With regard to the ladies of the harem, they fade out of 
the picture almost immediately, whereas in Pan Ku's version 
they seem to pervade the scene right down to the end. Prob- 
ably the divergence is more apparent than real, since both 
authors were so economical in their use of definitive subjects 
to their sentences. As for the holiday mood of these ladies, 
Chang Heng only just touches on it, and then goes straight 
to a more serious matter: they go and view the heaps of the 
slain and, as they return, turn back to take yet another look. 
It is hard to believe that the creator of the poor escaping 
pheasant and hare did not mean here to suggest a touch of 
sadism in these light-hearted young women. His sensitiveness 
to unconscious cruelty comes out again in relation to the fish- 
ing with a dragnet and its "womb-grabbing and egg-snatch- 
ing." And to this must be added an item in the final acrobatic 
scene in the entertainment: the sensational appeal he notes 
in the boys seeming to fall headlong and then being caught 
by their heels. We are brought face to face with that avid 
desire for violent sensation which characterized the gladi- 
atorial shows in imperial Rome, and which is so near to brutal 
blood-lust. 

Each thumbnail sketch has extraordinary precision of de- 
lineation. As part of this should be noted Chang Heng's 
emphasis on the underlying staff work to these large-scale 
festivities. There is the steward on a horse going round and 
supervizing the catering arrangements, there is — of all un- 
expected figures — the "traffic cop" {chin ch^e) marshalling 
the carriages in convenient parking places. Then there is the 
steward in charge of boats, with his oarswomen ready to break 
into a chanty. And then we are introduced to the inspector 

\ 128 1 



CHANG HENG ON THE WESTERN CAPITAL 

of marshes and fishing — reproved for breaking the very reg- 
ulations he was commissioned to enforce. 

On the other hand, along with this precision of detail goes 
a marked vagueness as to time. This was observable in Pan 
Ku's account, and it is clear that both authors aimed at de- 
picting a continuous round of pleasure-seeking in which night 
was turned into day. There is no need to assume that the 
Emperor and his circle engaged in all the diversions avail- 
able, and Chang Heng's story shows His Majesty taking 
his ease, in the daytime presumably, under his priceless chia i 
coverlets in the P4ng Lo lodge. If the entertainment and 
all were crowded into one day, the Emperor's rest must 
have been a very short one. One question arises in connection 
with the shooting. Did Chang Heng know that wild geese 
flew by night or at crack of dawn? From references to these 
birds in old poems we may assume that he did know. But 
then presumably the shooting party went straight from feast- 
ing to their sport, and the historian wonders whether their 
marksmanship was really as accurate as it is represented as 
being. 

What are we to make of Chang Heng's sudden outburst 
of indignation against the fishing party? It is a twofold out- 
burst, on account of the breaking of the law by the very 
people who should have known better, and on account of 
the thoughtless frivolity which brought about such wanton 
destruction of life. The historian's question here is whether 
there is not on both counts a distinction to be marked here 
between the ethos of the Former and Later Han dynasties. 
The Chou feudal society had realized the economic impor- 
tance of repressing the poverty-stricken peasants' lust for meat 
and fish, and there seems no reason for doubting that the 
former Han law code contained equally repressive measures. 
That, however, is a different matter from what emerges to 
view here. Chang Heng's indictment is of the court circle 
for assuming that they were above the law and for having 
no conscience about exterminating the spawn, etc. Since in 

[ 129 ] 



CRITIQUE 

that last century of Later Han the moral implications of 
high doctrine were only just beginning to win their way in 
the higher ranks of society, the strong inference is that Chang 
Heng's attitude represents one of the marked differences be- 
tween Former and Later Han. 

Coming to the great entertainment, we find Chang Heng 
giving it a name, ^'Chueh Ti^^ ("Horn-Butting") and launch- 
ing straightway into a detailed description from which horn- 
butting is singularly absent. That fact plus the strikingly 
entertaining characteristics of the show lead us to see what 
a radical change came to these old-time contests and mimetic 
dances. What originally were tribal rites, functioning as stim- 
ulants of tribal unity and joie de vivre^ now appear in new 
and elaborate forms as variety shows aimed at exciting the 
interest of a blase emperor and his court. The actual chueh ti 
contests clearly originated in Western Ch'in, as certain dances 
came from the Yii River in Szechuan (cf. Dubs' History of 
the Former Han Dynasty ^ vol. II, pp. 129-31). To these 
Emperor Wu's impresario and his successors added a num- 
ber of other items from all over the empire and its adjacent 
tribes. "Impresario" is the only term that does justice to the 
high managerial ability which could not only devise the more 
complicated tableaux but also bring them to the point of 
successful presentation. The significance of this lies in the 
fact that these impresarios took the place of the tribal sha- 
mans, and what originally had been shot through and through 
with the presuppositions of animistic religion were now 
adapted to a completely non-religious end. The process of 
change doubtless had begun before the Han era, but it was 
the wide expansion of the empire which brought matters to 
a head. To the Chinese conquerors the outlandish dances and 
mystery stunts they saw were not awe-inspiring but interest- 
ing, exciting, even amusing, in a word, of purely secular 
interest. 

Chang Heng himself is evidence of this secular temper. 
His account is solely concerned with reproducing the spec- 

[ 130 ] 



CHANG HENG ON THE WESTERN CAPITAL 

tacular effect. For him here was an elaborate specimen of 
the art of showmanships and the interesting thing is that 
he saw it as so elaborate that it defeated its own end. The 
Emperor in the last resort was not amused: he was bored. 
That, of course, is the force of the reference to the magician 
of the Eastern Ocean (the coastal area of the newly-settled 
dependency of Yueh in the far south). Just as Old Master 
Huang's magic powers proved powerless and he was eaten 
up by the very ravening beasts he claimed to control, so the 
marvelous succession of exciting shows failed to entertain: 
"they did not work." The translation hardly does justice to 
the homely force of the actual expression ^'fu shouj^^ literally 
"not saleable." If we may assume-^and it is but an assump- 
tion — that we see here a slang expression current in the shop- 
ping quarter of Loyang in Chang Heng's day, the poet's 
dramatic irony becomes yet more devastating. 

12. The Emferor^s Undignified Amour 

The final item in the entertainment would seem naturally 
to belong to the general account of it. Psychologically it is 
closely connected with the next phase of Chang Heng's story. 
In spite of the thrilling nature of this acrobatic display and 
the patriotic tenor of its finale, the Emperor's mounting feel- 
ing of ennui explodes into action. He cannot bear any more. 
He orders his carriage j and the court follows suit. The stately 
procession back to the palace proceeds. In the poem so far 
there is no indication of the emperor's frame of mind beyond 
the cryptic hint conveyed in the words "/^ shih fu shou^'^ 
("thus it was it did not work"). Then with lightning speed 
comes the denouement. In one pregnant triplet the Emper- 
or's boredom is depicted, and in one couplet more he is found 
creeping out of the palace incognito. 

The Son of Heaven, who throughout the major part of 
the poem has been chiefly a shadowy figure in the background, 
now comes right down stage and is subject to the full glare 
of the spotlight. Actually Chang Heng's language is im- 

[ 131 ] 



CRITIQUE 

personal as usual: only the sense indicates without question 
that he had an emperor in the forefront of his mind. Thus 
the first straight view we get of His Majesty is as a very 
human being subject to petulant longings for freedom of 
action. Since "the divinity that doth hedge a king" has so 
marked an emphasis in Han literature, Chang Heng's em- 
phasis on the other side is remarkably worth having. It is 
the more so because this author of ours, this poet-cum-scientist 
who belonged to a generation notable for its conventional 
prudery, gives a most unprudish description of a courtesan 
and her seductive tricks. He even says — not necessarily in 
a cynical but certainly in an ironic vein — that even a Buddhist 
monk could not but succumb to such charms. Approaching this 
matter from the historian's angle, it is to be noted that Em- 
peror Wu and, four reigns later. Emperor Ch^eng are on 
record as having made incognito expeditions, mention being 
made of viewing cock-fights and the like, to which items 
Chang Heng makes no reference. 

13. Final Conclusions 

So then, human nature being what it is, what follows? The 
poet outlines the resultant situation, stating the premises in 
blunt four-word couplets following on with six-word couplets 
embodying historical evidence, and so moving to trenchantly 
stated conclusions. His premises are two. The first is that 
a monarch may be expected to give rein to his carnal desires 
and pursue a course of ^^carfe diemP That is the force of 
the quotation from the "Odes" (see James Legge, tr., The 
Chinese Classics ^ IV, 176): "if you do not enjoy what you 
can now, other men will enjoy the possessions you discount." 
The second is that, given that temper and a man endowed 
with the authority to establish precedents, you cannot expect 
him to submit to conventional procedure (/i). 

Thus abruptly does our second-century court official, Chang 
Heng, bring us face to face with what is perhaps the most 
crucial problem of all monarchical forms of government. 

[ 132 ] 



CHANG HENG ON THE WESTERN CAPITAL 

Certainly it was so for the Han statesmen, engaged as they 
were in the first Chinese experiment in monarchy on the 
grand scale. From day to day, from year to year, and reign 
to reign, it was under the emperor's seal that precedent was 
created. That fact alone is enough to demonstrate that the 
issue of a constitution was up for consideration. And the fact 
that Chang Heng went straight to the issue of "/jo ku^^ ("the 
making of precedent") demonstrates his awareness of the 
problem. We have already approached this question. Now 
we can observe it from the angle of what Chang Heng had 
to say about the last generations of Former Han. He makes 
an indictment, citing two episodes, one in Emperor Ch'eng's 
reign, the other in Emperor Ai's. Emperor Ch^eng overrode 
all opposition and made a slave girl empress, creating a new 
rank in the harem in order to do so, and then gave way to 
a passion for her sister, ending by promising her that "none 
should rise above her," a pledge he kept by murdering two 
sons of his body by other women who might have risen to 
be mother of the heir-apparent. (For the whole unsavoury 
story see Dubs, II, 369-372.) Emperor Ai raised his male 
paramour Tung Hsien to the most exalted status among his 
subjects, making him rich beyond any other noble. Not only 
so: according to the record (see Han Shu, ch. 93) at a feast 
Emperor Ai once said with a laugh to Tung Hsien that he 
might follow Sage-King Yao's example and make him co- 
ruler and successor, as Yao did with the saintly Shun. 

This flagrant levity brought an instant protest and that, 
as Chang Heng notes, from a humble attendant. The record 
gives as his words: "The empire is the Founding Emperor's 
possession, not your Majesty's. It is of the highest importance 
that the Son of Heaven should not speak about his great 
heritage in jest." What Chang Heng does is to take the gist 
of that statement and use it as a springboard for his denunci- 
ation of the regime. 

With regard to Chang Heng's final count against Former 
Han, we are forcibly recalled to the deep-lying materialistic 

[ ^33 ] 



CRITIQUE 

vein of the supercilious nobleman. In profoundly cynical 
fashion he here enunciates the damning outcome of Han 
prosperity. True, Emperor Kao Tsu by his labors inaugurated 
a mighty kingdom. True, his immediate descendants built on 
his foundation and confirmed and extended his great heritage. 
But what followed? A temper of overweening arrogance, un- 
bounded extravagance, and irresponsible pleasure-seeking in 
the capital and the court. The fertile soil was theirs, the capital 
was safe from attack, why should anyone worry? So runs the 
poem — up to a point — then it changes. The farther a stream 
flows, the harder it is to damj the deeper a tree is rooted, 
the more difficult its uprooting. But these things do happen. 
"A fragrance which becomes more and more pungent ends 
in being nauseating." That is the poetic image, a superbly 
realistic one, not savagely satirical but imbued with tragic 
irony. The idea conveyed is of an inevitable doom, brought 
about not by gods or demons, but by the inner compulsion 
of events. 

Chang Heng's reaction to the situation is the more sig- 
nificant because of what he saw with his own eyes in Later 
Han, for example, in Emperor An's reign (a.d. 107-125). 
The boy-emperor was put through the rigors of a strict moral- 
istic education. As a young man he was thwarted by his all- 
powerful regent, Empress-Dowager Teng. After her death 
he was theoretically free to direct and rule. By then, however, 
the damage to his character was done, and this by the in- 
dulgence of the same kind of illicit impulses that Chang 
Heng describes here. Can it be doubted that the poet was 
here not thinking only of dead and gone emperors of the 
last regime? 

14. The Nobleman Makes His Bow 

The inexperienced reader may be led astray into thinking 
that the nobleman's disparaging words about himself either 
are not in keeping with his character, or are grossly hypo- 
critical. Neither is true. His apparent self-abasement is noth- 

[ 134 ] 



CHANG HENG ON THE WESTERN CAPITAL 

ing more than the use of the current coin of polite inter- 
course. But these stilted phrases should be read carefully. 
They show that in the interchange of debate formal homage 
was paid to the majesty of truth j and that in this sense, that 
a personal point of view was recognized as only too liable 
to be prejudiced, one-sided, and so misleading. A debater 
might, therefore, be supercilious and dogmatic, but he would, 
if he remembered his manners, have to make some such for- 
mal apology as this here. 

The illuminating remark is, of course, the one about hear- 
say, carrying as it does the implication that hearsay, com- 
pared with direct observation, is liable to be untrustworthy. 
There is ample evidence that Chang Heng was not the only 
scholar of his generation to be thus awake. Pan Piao had 
come very near to raising the problem, if he had not actually 
done so, by questioning the reliability of Ssu-ma Ch^en's 
"Record of History." Wang Cheung was writing his Lun 
Heng and deriding the cock-and-bull stories that were so 
widely believed. And the Ancient-Script scholars were pro- 
testing against the fabrications in the "Woof Books." The 
interesting thing is that this attitude of doubt is in relation 
to a period not more than three to one centuries back. We 
have to face the related fact that when it came to the far- 
back ages of antiquity both Chang Heng and Pan Ku, on 
the evidence of their ]u^ had a considerable bump of credulity. 
To all appearance they accepted the Sage-Kings as far back 
as Fu Hsi, shrouded though he was in the mists of antiquity. 
That being so, we must accept both states of consciousness 
as having carried on alongside each other. In some ways we 
should expect a critical mind to develop first with regard 
to more recent hearsay. There were much more complicated 
sets of information to be appraised, and they were not 
shrouded to the same degree in the nimbus of traditional 
religious belief. The historian bears in mind that in this age 
the claims of the central government to settle problems of 
belief were accepted in theory but not in practice. The thinkers 

[ 135 1 



CRITIQUE 

wanted to synthesize traditional and contemporary knowledge, 
old hearsay with new, but they could not agree on what was 
knowledge, what was proved fact. This kind of intellectual 
situation is familiar to historians of Greece and Rome, not 
to speak of mediaeval Christianity and the age in which Des- 
cartes subscribed his belief in the basic dogmas of the Church. 
The ability to think critically and empirically would seem to 
have been an uncertain growth, but that is no reason for 
minimizing the importance of its earlier phases. 

To come back to the nobleman and his last words, the 
emotional links in his mind are delineated with a fine sense 
of logic that brings his far-back past very near to our moder- 
nity. First, there is his exclamation over the country's un- 
paralleled wealth. Then, dyed-in-the-wool aristocrat that he 
is, he makes aristocracy's unconscious assumption that the 
main part of such wealth should be at its disposal. Finally, 
he bewails the fact that nowadays luxury, informed with ele- 
gance, is disallowed as not part of "the country's glory" 
among the nations. Then, with the inevitable gibe at these 
modern high-faluting principles of frugality, he winds up 
with a neat confession that naturally he would like to have 
the palmy days of the old regime back. 

Surely one has to acclaim the brilliance of Chang Heng's 
power of characterization. To go beyond literary criticism to 
the sterner call of history, that remark about wanting what 
one cannot get and not wanting, not being satisfied with, what 
one has, is highly reminiscent of Hsiin Ch^ng, the third- 
century B.C. philosopher who upheld ethical principles, but 
was deeply tinged with Legalism. He regarded human na- 
ture as insatiable in its desires: a man having reached his 
desired goal this year, next year pines for something more, 
and so on without end. Thus, he argued, the essence of gov- 
ernment consisted in keeping people as far as possible con- 
tented in that state of society in which their lot was cast. 
This entailed not only an iron hand in the government but 
also education of the people to habits of thought and action 

[ 136 1 



CHANG HENG ON THE WESTERN CAPITAL 

which embodied for them at their level the ideals of artistry 
and perfection. From that he went on to the Son of Heaven 
at the top of the human scale, set apart from all lesser mor- 
tals, but also ennobled by an environment in which beauty 
and order are displayed at every turn. As a theory empha- 
sizing the basic connection of art with ethics, this doctrine 
of Hslin Ch^ing's has its very great interest, and his influence 
can be traced through the Han era. Here, then, Chang Heng 
introduces the idea, not as a piece of philosophic theorizing 
but as a natural sentiment in the mouth of the nobleman. And 
the same lordly individual can cite the authority of the 
"Cricket's Song" to prove the laudability of his desire for 
luxury. But the "Cricket's Song," one of the T^ang Cycle 
of the "Odes," was a simple peasant ditty calling for one 
night's jollification in the midst of months of toil. Thus in 
one neat stroke Chang Heng pillories the cynicism that lurks 
behind the lordly viewpoint. 



[ 137 ] 



Chapter IX 



CRITIQUE OF PAN KU^S 
"FC7 ON THE EASTERN CAPITAL" 

<><c><><><X><>O<^<><>0<X^ 

I. The Eastern Capital Host Jibes at the 
Western Cafital Guest* s Ofinions 

From this point on, the author of the two fu is not making 
out a fictitious case but using the fersona of the man from 
Loyang to state his own point of view. It would appear that 
the critic has a simple psychological situation to examine: in 
a word, the stooge having had his say, we now get the truth — 
at least the truth as Pan Ku saw it. As we proceed, however, 
we shall find that the situation is not quite so simple as it looks. 
The statement about the effect of habitat and custom on 
a man's outlook has quite a modern dialectical ring to it. 
Nevertheless our first-century poet-historian was capable of 
it, and we know him well enough now not to be surprised 
at this. We may do well to remind ourselves that his family 
had for many generations been domiciled in the North-West, 
in the days before its leading members had obtained high 
office in the late Ch'ang-an court. Whether this increased 
the subtle range of Pan Ku's self-consciousness must remain 
doubtful, but in any case, here is a Loyang man twitting 
a Ch'ang-an man for being so provincially hidebound. The 
jibe is highly significant, for it is on the ground that the 
typical Ch'ang-an mentality has all along been tainted with 
the virus of the barbaric tyranny of Ch^n. That being so, 
the real virtus of "Great Han," the emergent significance 
of its words and deeds, is incomprehensible to a Ch^ang-an 
man. Now, if this sentiment had come from a second-rate, 
ill-informed mind, we might be warranted in discounting its 
significance. But it comes from Pan Ku, who had spent long 
years on the Loyang archives. We have the evidence of his 

[ 138 ] 



PAN KU ON THE EASTERN CAPITAL 

"History of the Former Han Dynasty" that he not only had 
very detailed knowledge of those two centuries but was well 
able to exercise his judgement in the matter. 

Assuming the high evidential value of his statements, we 
find them remarkably illuminating. First, there is the opinion 
that only a Later Han man, free from Ch'ang-an obsessions, 
is enlightened enough to appreciate the true worth of "Great 
Han." That, as subsequent sections will show, means that 
the high moral significance of the Han Dynasty only came 
to full expression with Han Redivivus under Emperors 
Kuang Wu and Ming (i.e. a.d. 25-75). What the full im- 
portance of this is will emerge later j but at this point this 
much is clear, that any attempt to run the two epochs to- 
gether, or even to equate the mentality of Former Han with 
that of Later Han, would have brought a protest from our 
author. 

Second, having in his Ch^ang-an fu surreptitiously pilloried 
Former Han from Emperor Wu down. Pan Ku here shows 
a different attitude to the founder of the line, Kao Tsu. He 
has a profound respect for him, so much so that he sees him 
as having broken the mould of history, as having initiated 
a regime which was unparalleled in previous ages. At the same 
time he maintains that the Ch^in virus was predominant in 
Kao Tsu's situation. Hence "the policy perforce" to establish 
the capital in the Kuan Chung area, where it would be stra- 
tegically safe, and to overawe contumacious elements by the 
magnificence of his headquarters. In other words, the found- 
ing Emperor, for all he had a commission from Heaven and 
fulfilled the wishes of the people, yet had no option but to 
set up a military autocracy. Lou Ching knew, he, the unknown 
exile who came out of the blue and so unaccountably was 
able to persuade the Emperor that Loyang would not doj 
and Chancellor Hsiao Ho, he also knew, and braved his mas- 
ter's anger for building the great Wei Yang Palace on the 
scale more or less of the ChSn Emperor's barbarous O Pang. 
And Pan Ku's point is that having started as a legalist autoc- 

[ 139 ] 



CRITIQUE 

racy, the Ch^ang-an government, for all its dazzling achieve- 
ments, was never able to rid itself of the Ch4n virus. The 
real answer can only be given in the light of Chang Heng's 
Loyang fu. 

Third, Kao Tsu's uniqueness consisted in part in the fact 
that he was a plebeian. According to Pan Ku there never 
had been before such a phenomenon as a plebeian ascending 
the throne. An odd statement this for two reasons. One is 
that in the versified eulogy of Emperor Kao Tsu in the "His- 
tory of the Former Han Dynasty" his family is represented 
(more or less convincingly) as tracing its descent back to the 
Sage-King Yao and having in Ch'u times become "Lord 
[kung] of Feng."^ The other is that according to the Yao Tien 
of the "History" Scripture {Shu Ching) and Mencius^ Sage- 
King Shun started life as a plebeian. Since it is impossible 
to assume that Pan Ku was ignorant of that eulogy, the high 
probability is that here in the poem he states the plain un- 
varnished truth (cf. Dubs, History of the Former Han Dy- 
nasty y I, 13-15). Likewise, it is impossible to assume that Pan 
Ku was unacquainted with the Yao Tien and Mencius. On 
the other hand, in the "Record of History" {Shih Chi^ ch. 13, 
init.) Shun appears as a descendant of the Yellow Emperor 
and so as a distant relative of Yao j so that Pan Ku may have 
had that version of the story in mind and, since Shun was a 
sage-king, have accepted it. On the two counts what emerges 

■^ This is apparently a reference to Emperor Kao's father, but no state- 
ment is made as to who raised him to this rank. In the Ti Chi (see Dubs, 
History of the Former Dynasty^ I, 40) the emperor-to-be becomes kung of 
P'ei by fiat of the more or less bogus King of Ch'u. Pan Ku's eulogy on 
Emperor Kao Tsu, on the other hand, is very studied in relation to the 
Emperor's ancestors, conveying the impression that the author, presumably 
Pan Piao, was not prepared to vouch for the truth of Fan Hsien-tzu's 
assertion about the family's aristocratic lineage (Dubs, of.cit.y I, 147). In 
the "Record of History" (ch. 8, fin.) a meeting of ministers refers to the 
just-dead Emperor Kao as being of "humble origin." As to the story of 
his miraculous conception either under the physical or psychological in- 
fluence of a dragon {Han Shu, ch. I, init.), may we not say that there 
the official propaganda was in the records, and Pan Ku's business was to 
transcribe it, whether he believed it or not? I come more and more to sus- 
pect that he did not really credit such stories. 

[ 140 ] 



PAN KU ON THE EASTERN CAPITAL 

to view is that in that highly class-conscious society of Han, 
both sage-kings and founding emperors had by hook or by 
crook to achieve noble lineage: a familiar phenomenon, of 
course, in other parts of the world. Further, Pan Ku's age 
was quite conversant with the spectacle of descendants of kings 
falling after five generations into the plebeian class. Emperor 
Kuang Wu was a notable example. 

Fourth, what are we to make of that remark about the 
Scriptures and sages? As an opinion it stands to reason, as 
any modern can see at once: if Kao Tsu inaugurated an un- 
precedented kind of era, then what he did was outside the 
cognizance of the Scriptures and sages. On the other hand. 
Pan Ku belonged to the first century, not the twentieth, and, 
in terms of what the prevalent school of thought in his day 
maintained, all dynastic history, past, present, and to come, 
followed an ineluctable pattern. Along that line of reasoning 
Han was just one link in the chain of causation which the 
Scriptures expounded and the sages foreshadowed. So then, 
as Pan Ku's mind formed this statement, it was functioning 
outside the bounds of current belief, and he was not afraid 
to show this. On both counts we get a side-light on that age, 
one in which new ideas had come to vigorous birth, and the 
Former Han type of teacher had to fight hard to retain his 
prestige at court. Pan Ku, unsure as he was of his future, 
could yet afford to write in this bold way — although, when 
Emperor Chang called on him to summarize those conserva- 
tive conclusions of the White Tiger Hall Conference, he 
could not afford to refuse. 

2. Wang Manges Usurpation and the Salvation 
of the Country 

The prosody in this section is irregular, the sentences not 
always running in strict pairs, and rhyme very noticeably 
absent at certain points. The argument is perfectly clear and 
moves forward by three main stages in well-marked para- 
graphs. The paragraphs deal with ( i ) the Wang Mang usur- 

[ 141 ] 



CRITIQUE 

pation and the reaction to it of Shang Ti (the Supreme Lord) j 
(2) Kuang Wu to the rescue j and (3) the moral beauty of 
the regime which he established. 

There is, on the surface at any rate, no cool appraisal in 
Pan Ku's attitude to Wang Mang. He condemns him utterly. 
In fact the sheer hyperbole of his indictment stirs the sus- 
picion that his words represent no more than the conventional 
court attitude of his day. This may disappoint those who are 
familiar with Dr. Hu Shih's and Dr. H.O.H. Stange's ef- 
forts to interpret Wang Mang favorably, their contention 
being that he made a serious attempt to put Confucius' teach- 
ing into effect and was defeated by the upper classes' violent 
opposition to the infringement of their privileges.^ A detailed 
examination of the Wang Mang regime is beyond the scope 
of this critique, but, assuming some measure of truth in this 
contention. Pan Ku's evidence is worth consideration. He in- 
dicted Wang Mang on the ground that his rule produced 
the most ghastly consequences throughout society. In effect 
the non-privileged classes suffered disaster equally with the 
privileged classes. As Pan Ku's appraisal of Wang Mang's 
life puts it (see Han Shu^ ch. 99C), "within the four seas 
there was the din of wailing, men lost the heart to rejoice 
in being alive." In justice to Pan Ku it must be recognized 
that to upset the equilibrium between the classes v/as just 
as opposed to Confucius' ways of thinking as neglect of the 
common man's welfare. There can be little doubt but that 
Wang Mang's economic policy was reckless in this as in cer- 
tain other respects. 

Moreover, Pan Ku's indictment contains the word "usurpa- 
tion," that is infraction of the principle of monarchic legiti- 
macy. It is very difficult for us to estimate the exact force 
of this word on a mid-Han man's lips. We are so familiar 
with the cycle of the sage-king dynasties as also with the fact 

^ Hu Shih, "Wang- Mang, the Socialist Emperor of Nineteen Centuries 
Ago," Journal of the North China Branch of the Royal A static Society^ Lix 
(1928), 218-230, and Hans O. H. Stange, Die Mono^rafhie iiher Wan^ 
Mangj Abhandlungen fiir die Kunde des Morgenlandes, Leipzig, 1938. 

[ 142 ] 



PAN KU ON THE EASTERN CAPITAL 

that "Great Han" eventually did go down, irrevocably, to 
be followed by other dynasties, each claiming a mandate from 
Heaven. We tend, therefore, to forget that for Pan Ku, being 
ignorant of later history, Han was unique, with its incom- 
parable virius destined to exist "for a myriad generations." 
Nor can the historian afford to ignore the possibility — or 
should I say probability? — that the theologizing imagination 
of courtier scholars dealt freely with the old myths and 
legends of far-back heroes with the very intent of substan- 
tiating the legitimacy of the Han throne. Compare, for ex- 
ample. Emperor Wu's faithful servants Ssu-ma T^an and 
Ssu-ma Ch^en in the earlier chapters of their "Record of 
History." There is this also to be borne in mind: a semi-tribal 
feudal order, as the Chou was, inevitably entailed divided 
loyalties j its history demonstrated this by a grim logic of 
events. But the Han monarchy, for all its semi-feudal struc- 
ture, called with ever increasing force for a supra-regional 
loyalty to "the One Man." Thus an unprecedented impor- 
tance came to be attached to the principle of legitimate suc- 
cession — as Kuang Wu's propaganda shows without a shadow 
of doubt and, the great tradition of ancestor worship being 
there to be utilized, usurpation became an unconscionably 
wicked crime, a dire offence against gods and men. 

The third paragraph is as hyperbolic in its encomium of 
the Kuang Wu regime as the second paragraph is in its de- 
nunciation of Wang Mang. The picture is significant for 
the very reason that it is so whole-heartedly tendentious. 
The facts are that Kuang Wu had first to give way to a 
more influential candidate, the Prince of Huai-yang, and after 
his death had for long years to fight hard for the throne 
he claimed on that "First Day" (a.d. 25). The situation ne- 
cessitated not only hard fighting but equally vigorous propa- 
ganda, and one of Kuang Wu's first acts was to appoint a 
group of fourteen Erudites. On grounds of local support his 
headquarters needed to be south of the Yellow River in a 
central location, so that Loyang was a natural choice j and, 

[ 143 ] 



CRITIQUE 

that being fixed, the Erudites could boost its virtus as the true 
capital. Loyang was associated with Duke Chou, the saintly 
sage of Confucius' adoration, so the propaganda had a rich 
field in which to spread itself. By what stages those Erudites 
developed the required mythological halo to their master's 
person is not clear from the records j and, since Pan Ku was 
writing twenty to thirty years after Kuang Wu's death and 
canonization, we cannot assume that the paean of praise here 
reflects the early propaganda. 

Certain ascriptions in this paean stand out as highly sig- 
nificant. First in importance is the repeated emphasis on 
Kuang Wu as the embodiment of consummate sageness. 
Claims are made here which surpass in magnitude anything 
ever claimed for the emperors of the Former Han dynasty. 
Thus Kuang Wu's merits and labors are represented as going 
beyond those of anyone of the sage-kings, and this view is 
clinched by examining various features of his career and per- 
sonality in the light of recent as well as ancient history. Thus 
Kao Tsu and his son. Emperor Wen, and the megalomaniac 
Emperor Wu come into the picture on the same footing as 
the three sage-kings mentioned. To the modern reader fa- 
miliar with the nimbus of semi-divinity attached in the Scrip- 
tures to the sage-kings, this freedom of comparison may 
appear extraordinary j but it is on a par with what is said 
about Kao Tsu and the Scriptures in Section i. One thing 
is clear: if the Former Han dynasty started moving in a theo- 
cratic direction, then the Later Han dynasty showed a defi- 
nitely more marked tendency in that direction. As for com- 
parison with contemporary Rome and its emperor worship, 
it is clear that Kuang Wu's formal apotheosis did not take 
place till after his death. 

Second, there is the striking phrase "Heaven and Earth 
were in revolution" {ke ming) : striking because of its apparent 
re-emphasis on theistic action. Again a theologically minded 
reader needs to beware of pitfalls. Ke ming occurs in the 
"Changes" Scripture, the scripture which in its appended 

[ 144 ] 



PAN KU ON THE EASTERN CAPITAL 

"wings" embodies a strongly naturalistic attitude to events. 
The passage comes in the Tuan ("Definitive Judgements") 
under the 49th Hexagram, the name of which was KS. Of 
the six illustrative definitions given there one reads as fol- 
lows: "Heaven and Earth ke [produce great changes] and 
the four seasons are completed." It is followed by T^ang and 
Wu ke ming [wrought revolution] obeying Heaven and 
fulfilling [the wishes of] men." Pan Ku in his argument 
also refers to T^ang and Wu, but as enacting Heaven's pun- 
ishments. Clearly the "Changes" passage was running in his 
head as he wrote, and he felt free to adapt the ideological 
formulae to suit the rhythm of his thought. 

Third, this paragraph on consummate kingliness has an- 
other link with the "Changes," namely the remarks about 
the Yellow Emperor as inventor of regional divisions, market 
days, boats and carriages, implements and weapons. We turn 
at once to the famous passage in the Hsi Tz^u ("Appended 
Judgements") of the "Changes." Comparison with that pas- 
sage reveals that there is no mention there of any sage in- 
venting "regional divisions," that Shen Nung is the inventor 
of market days, and that Yao and Shun are associated with 
the Yellow Emperor as responsible for the other inventions. 
What happened to Pan Ku, then, in this connection? Was 
his recollection of the passage inaccurate (there are numerous 
instances of such inaccuracy in later classical and postclassical 
literature), or did he in summarizing the passage feel free 
to treat the Scripture passage in this fashion? On the evi- 
dence of these two sections of his fu^ the latter alternative 
is more likely, and we have another straw in the wind show- 
ing that he set a little light to sacred myth and legend. After 
all, he was a litterateur and not an erudite, and evidence 
has yet to appear that he belonged to any contemporary school 
of dogmatic thought. 

The next stage in the argument is couched in terms of a 
profound moral renovation, all the basic relationships ex- 
cept that of friendship) re-created on a higher level of efficac}'. 

[ 145 ] 



CRITIQUE 

In fact the language about the relation of sovereign and sub- 
ject seems to indicate that Kuang Wu had built it up for 
the first time in history. This, however, cannot be Pan Ku's 
strict view-point, since he claims Fu Hsi as the originator of 
the idea of basic human ties. 

3. The Emferor Ming^s Achievements 

The narrative is extraordinarily condensed, passing quickly 
from a few couplets on Emperor Ming's initial procedure to 
his first royal progress, and from that to the enlarging and 
beautifying of the new capital. Every word, therefore, is 
significant. Take, for example, the six words ^'sheng San Yung 
chih shang i" ("developing the lofty procedure of the San 
Yung," i.e. the Ming T'ang, Ling T'ai, and Pi Yung) : they 
refer to most spectacular innovations of the new regime (see 
Critique of Section 9 and Chang Heng's detailed treatment). 
So also with the royal progress, the term "the myriad states" 
shows that Pan Ku had the remoter districts of the empire 
in mind as well as the old central states. So also with the 
building of Loyang, a vast and complicated undertaking: in 
contrast to the description of the building of Ch'ang-an it 
occupies only six couplets. Yet in that space Pan Ku strikes 
a new note in constructional art: the extravagant, the rococo 
are superseded by symmetry and restraint. So also with the 
laying out of the park: only three couplets are devoted to it, 
but stress is laid on its moderate size and the comfort and 
safety of its denizens. 

When all this is added up, checked over with the "History 
of the Later Han Dynasty," and viewed in the light of the 
claim that Emperor Ming's reign doubled the splendour of 
Kuang Wu's achievement. Pan Ku's aim becomes clear. The 
material glories of the old capital to which he had devoted 
pages and pages of vivid description in his Ch^ang-an ]u 
should be taken as "trifling accomplishments." The new capi- 
tal in contrast, dignified though it was in appearance, repre- 
sented a fundamentally different achievement, a spiritual one 

[ 146 ] 



PAN KU ON THE EASTERN CAPITAL 

which brings man at last into right relations with man. This, 
achieved in the first instance by Kuang Wu on the scale of 
the central states, is now accomplished by his filial son on 
the wider scale of world empire. Admitting that the earlier 
regime brought an amazing development of material civili- 
zation, this is nothing in comparison with the moral civiliza- 
tion which the new regime now offers to all mankind. 

The temptation is to regard all this as mere court flattery, 
rather ominous flattery in view of the later terrible decline 
in governmental morale. Some presage of that decline ap- 
pears in Chang Heng's fu written some twenty to thirty years 
later. Yet it is well to reiterate that in Pan Ku's day no one 
could know that that was to come, whilst, on the other hand, 
peace and prosperity were there on a very wide scale. The 
real point is, then, whether Pan Ku had good ground for 
honest enthusiasm over the court he knew. To this angle 
of enquiry there can be only one answer, namely, that he 
had adequate ground, and the more so since he saw with 
such clarity of vision how in the Han era his China had 
moved on to an unprecedented level of civilized living. To 
judge by his biography he wrote his fu in Emperor Chang's 
reign, at a time when the principle of the new era had not 
yet become tarnished. It was still the springtime of Han 
Redivivus, comparable to what Gilbert Murray has envisaged 
as happening in the history of ancient Greece, after the 
Armada and York Town: "exhilaration in the air, a sense 
of walking in new paths, of dawning hopes and untried pos- 
sibilities.'" 

4. The Ifnferial Hunting Preserve and the 
Seasonal Hunts 

In his ^^Fu on the Western Capital" Pan Ku used thirty- 
four couplets in describing the Ch'ang-an imperial preserve, 
and even more in describing the hunt itself. Here the ac- 
count of the park and the hunt is complete in thirty-odd 

^ The Legacy of Greece^ Oxford, 192 1, p. 13. 

[ 147 ] 



CRITIQUE 

couplets. The same poetic power is found at work, producing 
a vivid picture, sensational in some respects, but completely 
nonsensational in regard to the actual killing. This can only 
be taken as deliberate on Pan Ku's part, as also his emphasis 
on the hunt as a military pageant with large detachments of 
horse and foot moving to station according to schedule. So 
with the driving of the game and the marksmanships it was 
so expert that not a bird or beast had a chance to escape, yet 
beyond a certain quota they were allowed to live. In a word, 
what we are given is a picture of drastic reform, of the blood- 
lust controlled through the Emperor's direct teaching. 

As we digest this section, Pan Ku's objective becomes clear. 
It is to portray a court that has effectively broken with its 
semi-barbarous past, which is now truly civilized: in its recog- 
nition of fundamental human relations, its conduct of solemn 
sacrifice, its refinement of the arts, its stately but restrained 
architecture, even in its hunts which so easily could become 
brutal orgies. And the guiding principle of it all is /i, ritual, 
propriety, courtesy, the right observance, the right action 
done in the right spirit in the right way: li with its charac- 
teristic virtues of ching (reverence), chiek (moderation, re- 
straint), and jang (magnanimous concession of other people's 
rights). 

This is not to suggest that before this age religious and 
social ritual were not much in everybody's mind. That, of 
course, would be absurd. From the days of the "Odes" down 
to the last generations of the Former Han dynasty when 
the ritual experts began to study the philosophy underlying 
li, the evidence is as bulky as it is incontestable. What is 
meant is that if we take Pan Ku at his face value, the Loyang 
court envisaged the principle of // more clearly than any 
previous court had done. Kuang Wu's Heaven-given mandate 
was held to be based not only on his Han descent, but also 
on his fidelity to all the moral implications of //'. These went 
further than strict observance of time-honored rites. Li was 

I 148 1 



J 



PAN KU ON THE EASTERN CAPITAL 

taken to be a passion-subduing, soul-releasing way of life. It 
was directly associated with the Heaven-ordained "Five Ties," 
and being so it was held to operate on the plebeian as on the 
patrician level. As our study of this fu, and comparison of 
it with Chang Heng's proceeds, these and other aspects of 
the Later Han "// ckiao^^ (the principle and dogma of //) 
will come into clearer relief. 

It may seem to some readers that this emphasis on li is 
in direct contradiction to what has been said about an exhil- 
aration in the air, a walking in new paths. But there is no 
real ground for the idea so often met with in quite learned 
circles, that under the influence of Confucius the Chinese 
people gave themselves to the worship of the great god Pro- 
priety, so much so that they became ultra-conservative and 
incidentally acquired abnormal powers of hypocritical pre- 
tence. As comparative sociologists know so well, all societies, 
whether retarded or otherwise cannot continue to flourish 
without a hard core of religiously sanctioned observance pene- 
trating into every department of their life. Not only so: as 
public opinion reacts to the pressure of events, custom changes j 
but the chances are that it does so under the facade of pro- 
priety. Thus the term "//" in these fu conceals the introduc- 
tion of innovations as well as the maintenance of traditions. 

Two references in this section call for comment. One is 
the citation of the Wang Chih ("Royal Regulations," a chap- 
ter in the "Book of Rites") and the "Odes." The four hunt- 
ing songs specified disclose no more than a far-sighted attitude 
to the preservation of game. The same attitude appears in 
the "Royal Regulations," but a new note is struck, that of 
humane consideration for birds and beasts. 

The other reference is that to "a thousand chariots and 
ten-thousand horsemen." In the constant fighting with Huns 
and other nomads, the Han statesmen had to learn the lesson 
that trained bodies of mounted archers could by their speed 
and mobility of manoeuvre break up solid phalanxes of foot- 

[ H9 1 



CRITIQUE 

soldiers reinforced by groups of war chariots. The lesson was 
learnt, and here we see large bodies of cavalry being trained 
in precision of attack. 

5. All Nations Turn to Loyang 

The speed with which in terse three-word sentences Pan 
Ku races through these all-important state functions is at first 
immensely surprising. Had he nothing more to say on the 
Ming T^ang, Pi Yung and Ling T'ai? In fact he had, but 
he put it into lyrical poems at the end of this ju^ so the ques- 
tions arising in this connection will be dealt with there. The 
purpose of this omission of all but the barest essentials is 
clear. The poet-author would have nothing to distract atten- 
tion from his central object of contemplation, the virtus ^ the 
Sacred Person, the Son of Heaven exercising his sublime in- 
fluence impartially throughout China and the whole world. 

The high enthusiasm of the passage is unmistakable. Pan 
Ku might well feel enthusiastic. In a.d. 59 there had been a 
small-scale durbar j in a.d. 69 Ai Lao, in the far South-West 
had voluntarily submitted and been made a commandery. 
The empire, so sadly diminished during the interregnum, 
was steadily being restored, by punitive, it must be confessed, 
as well as peaceful, measures. Hence the almost mystical 
fervor of the phrasing, with the climax "no one on land or 
water but breathless and trembling comes running to pay 
homage." As a matter of cold fact Emperor Ming despatched 
several punitive expeditions during his reign. 

Two problems of general history call for notice. One is 
"Chinese universalism," a bemusing term in its very various 
contexts. It comes to view in the opening couplets, for they 
throw into clearer relief the sacrificial function of the throne. 
It is a function operating not only for the Chinese nation 
but for all peoples: the Son of Heaven is in principle high 
priest for all nations, for Heaven is one and the earth is 
one, and Heaven's regent activates this oneness by his medi- 
atory sacrifices and his calculations in his observatory. The 

[ 150 ] 



PAN KU ON THE EASTERN CAPITAL 

question is, whether this is not in fact the first time in the 
Chinese history of ideas for this universalist principle to be 
found clearly superimposed on traditional beliefs. This ques- 
tion is much too big and complicated for any sort of answer 
to be attempted here, but two sets of evidence may be ad- 
duced by way of clarifying the issue. 

(i) The Yin-Yang theorists were at work all through 
Former Han times, and their cosmological rationalizing had 
its effect on scholars of the main tradition. One result was 
in the field of the "Changes" scholarship, and there was the 
production of that section of the "Appended Judgements" 
in which Fu Hsi, first of the sage-kings and reputed as the 
inventor of the Eight Trigrams, was represented as discov- 
ering the essential relations of all phenomena above and be- 
low. By his discovery he was able to give true direction to 
human affairs. It was to this passage that Pan Ku's mind 
turned as he thought of his emperor's unique position in the 
world. He made the tremendous claim that the Son of 
Heaven here and now, not only is able to do the same, but 
actually is doing so. Ideologically speaking, this was a move- 
ment of the mind going well beyond any such claim for a 
mysterious sage-emperor shrouded in the mists of antiquity. 
As we saw in Sections i and 2, Pan Ku claimed complete 
sageness, more than any one sage-king had achieved, for 
Emperors Kuang Wu and Ming. 

(2) The recorded acts of Emperor Wu show that he tra- 
velled widely, going out to the frontiers of his empire, and 
that he sacrificed there to the local gods and gave orders for 
these sacrifices to be continued. He claimed that the gods 
concerned responded to his offerings. This would seem to 
be a first stage, leading to a syncretizing period in which 
foreign deities might be adopted into the Chinese pantheon. 
Modern scholarship tends to substantiate such a movement 
in Han times and evidence pointing in this direction is to 
be found in Yang Hsiung's Kan Ch'uan Fu and Chang Heng's 
Ssu Hsuan Fu. Also Pan Ku's special reference in this sec- 

[ 151 ] 



CRITIQUE 

tion to the hundred ling (deities) points to his having had 
foreign gods in mind. But his enthusiastic idealism here 
carries him well beyond the mere recognition of such prac- 
tices, and that dictated by political expediency. He envisages 
a world society brought into existence by a world savior, 
acting as priest at the hub of the universe. Again we are 
reminded of Virgil and his Caesar Augustus — to the myth- 
making impulse (if it may rightly so be called) which comes 
to poets stirred by the opening of new horizons. 

The other general problem is that in connection with the 
dividing line between religion and politics. Quite clearly in 
Pan Ku's mind there was no dividing line: religion and poli- 
tics were one. And yet they were not so in the crass uncon- 
scious way that they are to the tribesman in a "closed society," 
as Bergson put it. On the contrary Pan Ku was vitally aware 
of an "open society" of the world. 

With regard to the durbar. Pan Ku's phrasing recalls the 
Chou feudal procedure, in which attendance of the feuda- 
tories at court was required on the first day of a new king's 
opening year — actually Emperor Ming's durbar was held at 
the beginning of his second year. We find the world outlook 
again emphasized in striking fashion, for not only tribute 
but also statistical tables and reports are represented as forth- 
coming from "all within the four seas." As the ^vt chapters 
on the barbarians show in the "History of the Later Han 
Dynasty," there was in the mid-Han centuries an increasing 
practice of sending such reports prepared by resident officers 
in the outlying regions. Once a barbarian kingdom became a 
commandery, statistical tables were prepared with a view to 
taxation. 

The official record of the durbar in 59 is not lacking in 
detail, for example, that the barbarian kings assisted in the 
sacrifice to Heaven and Earth, a detail which Pan Ku omits. 
We are, however, indebted to him for the information in 
the vivid picture painted here: the formal reception of the 
Emperor's guests, the tents prepared for them in one of the 

[ 152 ] 



PAN KU ON THE EASTERN CAPITAL 

palace-enclosure courts, the rows of court-officers in attend- 
ance, the coaching of the strangers in court etiquette. The 
account reads like that of an eye-witness, and, as we know, 
Pan Ku was attached to the court at that time. As to the 
feast held in the courtyard, emphasis is laid on it being on 
a grand scale, but there is no actual statement that Emperor 
Ming himself was present. The reference to his "conviviality" 
{huan chieh) points to him having graced the occasion. But 
then why is it that "at the sound of a bell the officers retire" 
and no reference is made to the Emperor retiring? Perhaps 
he was not actually present, and his conviviality comes into 
the picture as part of that mystic influence that is so stressed 
in this section. In any case here in the feast is an instance 
of that eudaemonistic strain which in Chinese classical thought 
goes hand in hand with high political mysticism. As to why 
arrangements should have been made for the officers to re- 
tire. Pan Ku is silent. Presumably, since some of the tribes 
were known to be heavy drinkers, it was to show that there 
was no further need for the feasters to stand on ceremony. 
The picture is consistent right down to the music and danc- 
ing at the feast: barbarian musicians and actors providing 
their part of the entertainment as well as the court profes- 
sionals. That the imperial virtus inspired the strangers' per- 
formance may be regarded as dubious historically j it would 
seem to be a mere poet's conceit. As a matter of fact, it is 
exactly in harmony with Pan Ku's line of idealization in this 
section: it puts a logical crown to it. The interesting thing, 
however, is that the same idea occurs in the Po Hu T^ung 
(see Tjan, of.cit., p. 396) and with the use of the same 
terms te {virtus) and kuang (wide-extending). So, unless 
Pan Ku, chief author of the Po Hu T^ung^ is to be accused 
of slipping his own ideas into that report, the "poet's con- 
ceit" was what the majority of the a.d. 79 conference thought 
about the matter. One of the alternative titles to the book is 
Po Hu T'ung Te Lun ("White Tiger Discussion on Widely- 
circulating Virtus^^). 

[ 153 ] 



CRITIQUE 

It is worth comparing two outlooks, Pan Ku's here and 
that of the "Music Record" {Yueh Chi) of the "Record of 
Rites." In the latter the purview is of one state and nation 
vis-a-vis Heaven and Earth. In our poem here music and 
dancing are something common to all peoples, a potential 
force for creating concord among nations, not merely between 
discordant elements in one community. As Paul Hindemith 
has said, "People who make music together cannot be enemies, 
at least not while the music lasts."* 

6. The Joyful State of Sober Simflicity and Earnest 
Endeavour, with Schools Everywhere 

Noticing that reiterated emphasis on "sageness" (or should 
it be "sanctity") in the man on the throne, and the universal 
feeling of delight which accompanies its untrammeled exer- 
cise, we are prepared for the idealistic picture the author's 
imagination created. What he presents is a Utopian society 
with its officials all under clear direction from above and 
themselves models of public and private frugality, with the 
Emperor and his court turning their backs on luxury. He 
seems a little naive here, for his purview is "within the four 
seas": as he knew well enough, great areas of the western 
and north-western empire were not suited to agriculture. 
However, that was not so in vast stretches of the southern 
empire. 

The picture is by no means wholly idealized. Thus, al- 
though the "Acts of the Emperors" {Pen Chi) do not record 
any very spectacular series of edicts on frugality, yet various 
items crop up there and elsewhere which show that Em- 
perors Kuang Wu and Ming had the matter well in hand. 
For instance, with regard to extravagance over funerals, in 
Kuang Wu's 7th and loth year he forbade excessive expendi- 
ture in this way, and Ming in his 12th year reiterated the 
command. Going further afield, we find at the end of the 
two chapters on Carriages and Robes in the "History of the 

* // Composer's Worlds Harvard University Press, 1952, p. 218. 

[ 154 ] 



PAN KU ON THE EASTERN CAPITAL 

Later Han Dynasty" a reference to prohibition of extrava- 
gance. In the introduction to the lives of the emperors, Kuang 
Wu is stated to have "cut out elaboration and studied sim- 
plicity." In contrast to the "three thousand occupants and 
fourteen grades" in the harem after Emperor Yuan's time, 
he had only three grades, the Empress and two ranks of 
concubines of good character. Emperor Ming is stated to have 
followed his father's example. In Emperor Chang's time, 
however, came a gradual decline in the morale of the harem. 
As for sumptuary regulations, the biographies of Kuang Wu's 
eleven sons (Hou Han Shuj ch. 72) clearly indicate that 
state officials could and did take action against princely of- 
fenders, whilst Prince Ts^ang did notable work as a watchdog, 
even remonstrating with his half-brother. Emperor Ming, 
for entertaining the princes too lavishly. 

As for the lower orders and the officials whose standard 
of honor and efficiency made so much difference to their lives, 
the general impression, gained from the Acts of the two 
emperors and the introduction of the "Biographies of Up- 
right Officials," is that with the establishment of peace the 
country quickly recovered j also that the duties of the various 
types of officials were pretty clearly marked, and they them- 
selves had every inducement to work for popular confidence. 
In a word, the new broom swept cleanly enough. Adding 
all these corroborations together, the safe conclusion may be 
drawn that Pan Ku had some solid basis of fact to his lyrical 
outburst. To this may be added the quite likely assumption 
that, writing as he did in Emperor Chang's reign, he had 
seen signs of an increase in luxurious living and aimed at 
probing the imperial conscience. This could most forcibly 
and adroitly be done by painting a picture of the ideal. 

What then about these moral ideals with the tale of 
which a Loyang man could override the upholder of old 
Ch'ang-an? The first thing to be noticed is that they are 
generally not so much Confucian as Taoist, and in the first 
instance not so much Taoist as Mohist. The devotees of the 

[ 155 ] 



CRITIQUE 

Tao never advocated hard work as the scholars and the 
Mohists did, and, for that matter, the Legalists j whilst the 
main bulk of the scholars had no cult of austerity even for 
the common people. The surprising thing is that Pan Ku's 
mind went straight from Mohistic emphasis to the extreme 
Taoistic ^^hsing shen chi mo^^ ("body and soul untrammeled" : 
Margoulies, ^^ concentres en eux memes^'^)^ and from that to 
the aristocratic scholar's metaphor of "sheen of jade and ring 
of gold." The passage as a whole is a perfect amalgam of 
the three schools, as such deserving the closest study j and 
this the more because later more dogma-entrenched ages took 
Mohism to have been dead by this time and Taoism basically 
incompatible with Confucianism. 

The obvious conclusion is that what may be true of later 
ages is not necessarily true of the '90's in Later-Han China. 
Rather, in that lively and optimistic age active minds were 
definitely stimulated by all these heterogeneous, doctrinal 
influences. As with the vast stretches of empire, both actual 
and possible now revealed, so with Chuang Chou's playing 
with a space-time infinity and his exquisite correlation of 
the individual with that infinity, these reflections were ex- 
citing intellectual experiences. Pan Ku shows us this here, 
and Chang Heng shows it in his Ssu Hsiian Fu when, having 
made his imaginary flight round the world and beyond into 
infinite space, he alighted in his ancestral village and found 
he could compose the quarrels of orthodoxy and Taoism and 
Mohism. Compare also the critique to the Mohist section 
of Liu Hsin's "Catalogue." The entertainment of veterans 
is there directly attributed to Mohist influence. 

The importance of the philosophy of education presented 
in this section lies in two directions. One is that although 
we hear nothing of Hsiin Tzu during the early Han reigns, 
when we come to the exaltation of the scholar official by 
Emperor Wu and his successors, Hsiin Tzu's teachings on 
ritual appear suddenly to have been found full of signifi- 

[ 156 ] 



PAN KU ON THE EASTERN CAPITAL 

cance. In the "Record of Rites" his writings are to be found 
quoted at length (although always anonymously). The nat- 
ural inference is that serious-minded provincial officials found 
that here was a potent force for "renovating the people" in 
their charge. The other direction is in relation to Pan Ku's 
generation and his own marked adhesion to Hsun Tzu's 
principles. The Po Hu T^ung (see Tjan, of.ciL, pp. 48-68) 
delivers judgement on schools in the provinces in the same 
vein. So the old guard of the Modern-Script party were in 
agreement on this score. As for Pan Ku, he saw this as the 
main lever for bringing the whole world of the barbarian 
peoples into civilized unity of life. Now let us examine what 
he actually says. 

Having explored the effect of imperial example on char- 
acter, Pan Ku's mind turned naturally enough to formal edu- 
cation. Here again his sentiments are surprising. First, he has 
nothing to say on the fruits of high scholastic training, noth- 
ing on the benefits of scripture knowledge: that is left to 
the next section of the poem, and there he is highly critical 
of the results. Here his admiration is confined to the fruits 
of village education. Second, his "forest of schools" as ap- 
plied to all "within the four seas" is a gross exaggeration, 
though not without some substance. Wherever settled gov- 
ernment came to prevail, there schools were set up by the 
regional officials without delay. It paid them to do so. 

There is a striking correspondence between the spirit of 
this passage and the teachings of Hsiin Tzu — that intelligent 
psychologist of the third century b.c. about whom later ages 
have had such doubts as to whether he may be counted 
a true Confucianist. Hsiin Tzu, having discerned in human 
nature an unslakable appetite for grabbing, urged the neces- 
sity for not only a stern government penalizing offenders 
against the common weal, but also the building up of an 
acquired self, one in which habits of mutual consideration 
came into effective action. The means to this end were family 

[ 157 ] 



CRITIQUE 

and community rituals including songs and dances.^ The so- 
ciologist has but to contemplate the daily, monthly, yearly 
repetition of such emotive, indeed poetic, formulas of speech 
and action to realize at once their tremendous social poten- 
tialities. 

School buildings are the setting to the picture, buildings 
devoted to education, in the government centres and also 
in the subsidiary village centres. In these buildings gatherings 
took place not only of teachers and scholars as such but also 
of plain folk, high and low, rich and poor, young and old, 
and this anywhere in the empire. The occasion is not speci- 
fied, but the reference to ritual implements suggests some 
traditional fertility rites to be performed j and the implements 
"being in full supply" might well indicate the help of Chi- 
nese officials directed to making the rite more impressive, 
more orderly, and incidentally more Chinese. While the 
dancing still embodied the old animistic mana, Pan Ku's 
"singing the songs of jen^"^ tells a different tale. A communal 
feast follows at which mutual deference is the ruling spirit. 
This is portrayed by a masterly image, ^Heng chiang yao yen 
chih /i," literally, "the feast formalities of [backs] going up 
and down." One wonders whether Pan Ku and his courtly 
readers smiled as they envisaged the peasants and their clumsy 
bows, their mouths full of concord — and their gusty sighs. 
The virtus would be the theme of high converse, the virtus 
which inspires hard work and courteous cooperation, and so 
brings peace and plenty. 

It all sounds like a rhapsodical dream, unrelated to real 
life. Yet in this greater Han era something happened in that 
vast area east of the Himalayas. When Han collapsed and 
the virtus myth seemed exploded for all time, that something 
proved tenacious of life through long centuries of upheaval 
and barbarian invasion, and eventually the empire came to- 
gether again. The foundations of Sinicization had been well 

* Cf. Fung Yu-lan, A History of Chinese Philosofhy^ tr. D. Bodde, 
Princeton University Press, 1952-53, I, 12, especially par. 7. 

[ 158 ] 



PAN KU ON THE EASTERN CAPITAL 

and truly laid at peasant level, old mores impregnated with 
a more civilized ethos. In Section 5 the critic has urged that 
the term "Han Confucianism" is a kaleidoscopic term, and 
opaque at that. However true that is, the "something" we 
descry at work was linked with that mysterious figure of the 
sixth century b.c, Confucius of Lu, over whose message the 
mid-Han scholars quarreled so violently. In the Analects y 
in that highly hagiographical section, the Tenth Book,^ there 
is this revealed: "When the villagers drank wine, the Master 
withdrew only after the aged withdrew, and when the vil- 
lagers did their rite against pestilence, he put on his court 
robes and stood to attention." 

7. The Eastern Hosfs Final Argument 

Here is the final demolition of the Western Guest's case. 
The prosody gives a hammering effect to the subject matter, 
most of the clauses being of four-word length. Of the eight 
indictments made, most are summarizations of what has been 
argued earlier in the fu^ and the critique need not restress 
them here. This, however, does not apply to the first indict- 
ment to which this section gives a very intriguing, indeed 
surprising, setting. There is also a curious quirk to the state- 
ment about the River Chart and Script, whilst Pan Ku's 
theory of knowledge, here stated with additional precision, 
requires scrutiny. 

As the host at the beginning of the fu accused his guest 
of superficiality of knowledge, so here alsoj but this time 
he couples him with those "who nowadays engage in dis- 
cussion." It is difficult at first to see why Pan Ku drags them 
in, for his Ch'ang-an fu consisted throughout of an appeal 
to facts. However, the connection in Pan Ku's mind is clear 
from his remark about the guests being "vainly carried away 
by non-essentials." The fault in both cases is that they have 
their heads buried in books, so that they have no eye for 

^ Assumed by some critics to be a later strain in the traditions about 
Confucius. 

[ 159 ] 



CRITIQUE 

essential facts, cannot make vital comparisons between the 
past and the present, cannot trace the real reason why the 
Han virti/^ has come to be what it now is. The accusation is 
not that knowledge of the Scriptures is a bad thing, but that 
absorption in that one kind of knowledge leads to the mind 
being bemused with trifles — an accusation to which certain 
remarks in the Christian Gospels about "tithing mint and cum- 
min" lends point in the Western mind. Fortunately we have 
right-to-hand a document which highlights this accusation. 
It is the Po Hu T^ung. Of the three hundred and twelve 
subjects dealt with there a very large proportion reveals the 
meticulous mind at work on trifles, whilst at every point 
proof is advanced in the shape of scripture citation. We may 
surmise that the making of this digest of the Conference's 
conclusions left Pan Ku with a feeling of irritation, so that 
when he wrote this passage in the fu his temper got the bet- 
ter of him. 

However we may explain this blunt attack on the scripture- 
mongers of the age, it is highly revealing in more ways than 
one. In the first place, the net result of Sinological studies 
on Han is that once the Scholars got into the saddle, they 
showed a markedly authoritarian temper of mind: everyone 
had to subscribe to belief in the sacred Scriptures. Although 
Pan Ku's outbreak substantiates the idea to a considerable 
extent, yet the intellectual situation was far from being mono- 
lithic. As we have already seen, it was a vastly contentious 
age with innumerable different recensions to the texts of the 
Scriptures, and every famous teacher fighting for recognition 
of his particular brand of orthodoxy. True, the Emperor was 
the "Defender of the Faith," and the faith was the defender 
of him, and this naturally induced a dogma-loving frame of 
mindj but it did not follow that the palace and the Pi Yung 
(where the Erudites were) saw eye to eye over the question 
of what dogmas should prevail. 

It is here that Pan Ku's outbreak stimulates the right ques- 
tions. Consider his personal situation. With his "History of 

[ i6o ] 



PAN KU ON THE EASTERN CAPITAL 

the Former Han Dynasty" practically finished he was anxious 
to get an administrative post, to get his foot on the first rung 
and so qualify for a high position in the official hierarchy. Is 
it likely that he would wreck his chances by saying things 
which he knew would cause odium in high places? Moreover, 
we know that his two fu poems received enthusiastic acclama- 
tion at court. That being so, we look for light on Emperor 
Chang and his intimates. The historians are agreed that he 
was not a notably masterful person, but that he was intelli- 
gent, with an enlightened interest in the more liberal views 
about the Scriptures. Why he felt compelled to give way to 
the conservative forces in the White Tiger Conference is not 
clear. What is clear is that he had a great affection and re- 
spect for his gentleman-in-waiting, Chia K'uei. Now, as his 
biography shows, Chia K'uei had worked with Pan Ku in 
the archives at the palace, and was a staunch, although not 
extreme, member of the Ancient-Script party. Also this party, 
weak as it was in Scholar circles and hated by the Erudites 
as a body, yet from Kuang Wu's reign on had influence in 
the palace (see Hou Han Shuj ch. 66^ in which is given the 
biographies of the party's outstanding members). We find 
one Ch'eng Yuan in a memorial to Kuang Wu expressing 
much the same sentiments about myopic scholarship as Pan 
Ku does. Since there can be no doubt but that Emperor 
Chang's sympathies lay with this party, we have a clue to 
the problem which the poem's blunt denunciation constitutes. 
We are also encouraged in the belief that the Later Han 
court was not so hide-bound as has generally been supposed. 
Turning to the River Diagram {Ho T^u) and Lo River 
Script {Lo Shu) J we find this enlightened author of ours 
stressing not only the natural advantages of the Loyang site 
but also its cabalistic virtues. That his words come through 
the lips of a poetically conceived Loyang host is not evidence 
that his creator either had or had not any doubts about the 
historicity of the two legends. The Chart was supposed to 
have been on the back of a dragon which in ancient times 

[ i6i ] 



CRITIQUE 

came to the surface of the Yellow River and to have con- 
tained Fu Hsi's original Eight Trigrams. The Script was 
supposed to have been on the back of a turtle, which emerged 
from the Lo River in Sage-King Yii's time, and to have 
consisted of a mystic pattern of numbers from one to nine 
connected with the science of government. Now the peculi- 
arity about the reference to them here lies in their having 
also been used in the Ch^ang-an fu to demonstrate that the 
Western Capital was the Heaven-ordained spot. The fact 
that such legends could be attached to two very different 
parts of the Yellow River will not surprise the comparative 
student of legend and myth, but it seems odd that Pan Ku 
should have allowed the two statements to stand in mutual 
contradiction. Again we face the possibility, both with Pan 
Ku and Chang Heng, that although they believed the old 
traditions which were universally held to be historically true, 
they accepted them with a grain of salt. In any case. Pan 
Ku was not sufficiently interested to clear up the confusion 
between the two stories. 

In sharp distinction to this nonchalant attitude stands his 
contrasting the worship of very minor deities {hsien) at the 
Chien Chang and Kan Chilian palaces with the enlightened 
procedures in the Ming T^ang, Ling T'ai, and Pi Yung. How, 
he asks, can they be compared in basic efficacy, in the uniting 
of Heaven and man? His underlying motif, both here and 
in the preceding sentence, and most clearly of all in the final 
two couplets, is that old Han in comparison with new Han 
was ignorant, benighted. Even though the Scripture obscur- 
antists do not realize it, a new day has dawned with the 
light of a new knowledge, a knowledge based on reason and 
giving full weight to spiritual values, in particular the world- 
uniting virtus. 

Pan Ku believed that the founders of the new order. Em- 
perors Kuang Wu and Ming, had reviewed whatever was 
evil and whatever was latently good in the old order and 
had eradicated the one and brought the other into full play. 

[ 162 ] 



PAN KU ON THE EASTERN CAPITAL 

At the same time Pan Ku believed that the various principles 
of this revolution had been envisaged by one sage-king and 
another. He did not therefore hold that Later Han wrote 
on a clean slate, but he did maintain that by the implemen- 
tation of those principles their full ethical grandeur had for 
the first time emerged to view. That he idealized can hardly 
be doubted. That there was some truth in his stated belief 
also can hardly be doubted. 

8. The Guest^s Discomfiture 

Here is another example of Pan Ku's sense of humor at 
work. As was seen in relation to those endless stairs of the 
Well-Head Tower, he was a master hand at envisaging and 
depicting the ludicrous aspects of a trying physical situation. 
It did not occur to him, as it occurred to Chang Heng, that 
the man who had the tower built was acting in absurd fashion, 
trying to pull himself up to Heaven by his bootstraps. What 
appealed to Pan Ku was the stumbler on the stairs in a sweat 
of nervous apprehension. So here the humor is more of the 
slapstick kind, not in the vein of polished wit. The poor dis- 
comfited provincial is ruthlessly depicted in all the confusion 
of a man who has made a fool of himself. As a side-light 
on those much vaunted manners at the Loyang court, the 
passage is revealing enough. And we can well imagine Chang 
Heng, with his mordant vein of sarcasm, reading it over and 
feeling it a little simple and heavy-handed, particularly as 
the new regime was not quite so perfect as the poem pre- 
sented it to be. 

9. The Five Odes 

No literary critic, whether Chinese or foreigner, will ques- 
tion Pan Ku's good dramatic judgement in relegating these 
lyrical outbursts to an addendum. They will also appreciate 
the skill with which he works them into the final denoue- 
ment of the guest's repentance. On the other hand, the odes 
themselves are so conventional, in phrasing as in thought, 
that the modern reader finds difficulty in seeing why they 

[ 163 ] 



CRITIQUE 

should have been regarded as having such a soul-searching 
effect. To that query the only answer is that since the fu 
received great acclamation, Pan Ku's sense of his public must 
have been right. His court readers felt that he had said the 
right thing in the right distinguished lyrical way. 

There is one underlying theme to the five odes, namely, 
the transcendental evidence which is forthcoming to prove 
that the new regime had divine approval. In the case of the 
Precious Tripods and White Pheasants these confirmations 
come right out of the blue. No specific action could bring 
their presence. Either true kingship was there, and they came, 
or it was not there, and they did not come. The Ming T^ang, 
Pi Yung, and Ling T^ai plainly are in a different category. 
In their case edifices were erected, and carefully planned 
rites and learned procedures carried out, these elaborate means 
being adopted to achieve one great end — symbolization of 
true kingship and proof thereby to gods and men that the 
occupants of the throne had both a legitimate and moral claim 
to be Sons of Heaven. That being so, the main question of 
this book, the difference between Former and Later Han, 
confronts us in relation to these three institutions. In what 
ways precisely were these rites and procedures innovations? 

The "Treatise on Sacrifice" {Hou Han Shu, Chi Ssu Chih) 
states that Kuang Wu in the last two years of his reign began 
to build a ming fang, a ling t^ai, and a fi yung, and that 
Emperor Ming in his second year in the first month sacri- 
ficed in this ming fang to the Five Shang Ti, proceeded 
thereafter to his ling fai, in the third month carried out a 
rite of archery in his p yung, and in the tenth month enter- 
tained the veterans of the state service there. As far as offi- 
cial evidence of specific Han institutions takes us, this is the 
first occasion on which such a combination of rites was per- 
formed in a Han capital, the first occasion on which all three 
centres were placed near to each other, each being furnished 
with a surrounding moat of running water. So much is clear 
to start with. It is also clear from the "History of the Later 

[ 164 ] 



PAN KU ON THE EASTERN CAPITAL 

Han Dynasty" (Hou Han Shu, ch. i8) that there was an- 
other ming t^ang at Wen Shang near to Mount T^ai, and in 
that area another ling t^ai associated with Sage-King Yao's 
reputed grave. No other fi yung site is mentioned. 

With regard to various ming t^ang in the past, the "History 
of the Former Han Dynasty" is highly instructive, particu- 
larly in relation to Emperor Wu. At his accession there were 
certain influential scholars who urged him to construct a 
ming t^ang, presumably in Ch^ang-anj but there is no infor- 
mation as to what function it should perform. The suggestion 
was quashed by the Empress Dowager, and there is no record 
of Emperor Wu having thought of it again. But in no b.c. 
on his first visit to Mount T'ai he discovered a ruined build- 
ing at Wen Shang in that neighborhood and learnt on en- 
quiry that it had been used by the Chou kings when on a 
royal progress. The Emperor turned to the scholars to tell 
him what form it had. They could not tell him, but a South- 
Ch4 shaman produced a plan dating, he averred, from the 
Yellow Emperor's time. (See Han Shu, ch. 25, where in 
considerable detail is described the cabalistic plan on which 
Emperor Wu built at Wen Shang a ming fang in 109 b.c.) 
This plan with its circular moat of running water, its square 
structural features et cetera, was at least on much the same 
plan as that which Emperors Kuang Wu and Ming used 
later in constructing their ming t^ang. This was particularly 
so in relation to the Five Ti (High Gods), for both ming 
fang contained five altars, one to each Ti, ranged north, 
south, east, west, and centre. 109 b.c. is the first recorded 
occasion on which the Five Ti were worshipped in a ming 
fang. It would seem doubtful, to say the least, whether the 
scholars of the old Lu State (in the vicinity of Mount T^ai) 
had any tradition which corresponded to what Emperor Wu 
had in mind on this occasion, and the same would seem to 
apply to the scholars concerned in the 140 b.c. move in Ch^ang- 
an. As to Emperor Wu's purpose in the matter, it is plain 
enough from his pronouncements at the time: he wanted 

[ 165 ] 



CRITIQUE 

everybody in China and out to know that he had the special 
favor of all the regional deities. 

Pan Ku's ode refers not only to the Five Ti, but also to "under 
universal heaven, from end to end of earth, each [dignitary 
assisting] according to his office" ("^o i chH chih^'*). In the 
"Filial Piety Scripture" (ch. 9) we find this statement: "In 
times past Duke Chou did a chiao [suburban] sacrifice to Hou 
Chi, making him f^ei ["associate, ancillary deity"] to Heaven, 
and an ancestral sacrifice to King Wen in the ming t^ang^ 
making him f^ei to Shang Ti [one or more High Gods], and 
herewith every prince and dignitary within the four seas, 
according to his office came and [helped in] the sacrifice." 
A very remarkable passage. The point here is that Pan Ku 
clearly had this form of words running in his head, and that 
the obvious meaning of ming t^ang in such a context is "ances- 
tral fane." According to Ts^ai Yung's (132-172) theory in 
his "Discussion on Ming T'ang" (cited by Liu Chao in his 
comment on Hou Han Shu^ ch. 18), ''ming fang^'^ was the 
name given by the Chou royal house to their ancestral fanej 
and in Six Dynasties' and T'ang times this seems to have 
been generally assumed. Thus it appears that there was an- 
other kind of ming fang embodied in tradition, one associ- 
ated with ancestor worship and filial piety. 

Yet another functional kind of ming fang appears in the 
Wang Chih and Ming T'ang Wei of the "Record of Rites," 
namely a building erected in Loyang by Duke Chou, in which 
he lectured the assembled feudatories on their duties, pub- 
lishing policy and conferring offices. The ruined ming fang 
near Mount T^ai would appear to have been an offshoot of 
this, namely a place of assembly for regional dignitaries when 
a Chou king was on progress (cf. Mencius I B 5). These 
m^ing fang on the face of it had nothing to do with sacri- 
ficing to the Five Ti. 

On the top of all this come the questions arising out of 
Pan Ku's "Five Ti." Where did they come from? Fortu- 
nately for historians Pan Ku himself supplies the key docu- 

[ 166 ] 



PAN KU ON THE EASTERN CAPITAL 

ment in that connection. In his monograph on the sacrifices 
in the suburbs (Han Shu, 25) one of his major concerns is 
to trace the rise of these several sacrifices in ancient Chou 
times. As the story goes on step by step, he engages confi- 
dence by his warning that he cannot guarantee the truth of 
the information he gives. Bearing in mind that "//" was a 
Yin-Chou term, particular as to meaning ("sovereign," "god") 
yet capable of becoming generic, the story can be outlined 
as follows. In the great western area where the Chou and 
Ch'in culture interpenetrated, certain Ch'in princes created 
over a long term of years a number of high places, at which 
sacrifices were offered to a White Ti, a Red Ti, a Green Ti, 
and a Black Ti. Then came the First Ch4n Emperor, and 
he seems to have rationalized these several worships by re- 
lating them to the natural-law symbolism of the Tsou Yen 
school of thought and the rotation of the four seasons, under 
the influence of wood, fire, metal, water, and earth. The 
political aim behind this is clear enough. Emperor Kao went 
a step further and sacrificed to Five Ti, adding a Yellow Ti, 
symbolic of earth. These Five Ti were regularly worshipped 
at one high place, at the juncture of the Rivers Wei and Pa. 
Then another god appears on the scene, a T^ai I (Supreme 
One), with sidereal connections, to whom the Five Ti were 
made ancillary gods {f^ei)J Emperor Wu, among his other 

"^ I use the term "ancillary" for two reasons. One is that "p'ei" at this 
time was commonly in use to denote the functional status of the wife to 
her husband. She was his mate, devoted to his interests, having a share in 
his reputation, but always subservient to him, her lord and master. "Mate" 
would make an excellent translation, but its modern vernacular usage rules 
it out as having too much an equalitarian connotation. What exactly the 
status of this relationship was in the list of the "Five Ties" (lun) is difficult 
to say. In some passages it figures high, being placed next to the prime 
relationship of father and son, in other passages low, although always above 
"friend and friend." 

The other reason is that f^ei is the term used in the "King Wen" decade 
of the "Odes" to denote King Wen's status in Heaven. That takes us back 
to the largely legendary beginnings of the Chou feudal order. I can find 
no evidence that f^ei was used by the Shang-Yin Kings in similar circum- 
stances, and it seems likely that Duke Chou initiated this usage. Since 
Loyang was the city whose first building was associated with Duke Chou, 

[ 167 ] 



CRITIQUE 

sacrificial innovations, had T^ai I's altar erected at Kan Chilian 
where the new summer palace was. This sacrifice continued 
throughout the succeeding reigns, as also did those to the 
Five Ti at the Wei-Pa high place/ Gradually the feeling 
arose at court that all these sacrifices should be concentrated 
in the capital and a ming t^ang be built there. The decision 
was taken, then reversed, then taken again, then again re- 
versed. There is no record in the "History of the Former 
Han Dynasty" of a completed ming fang at Ch^ang-an until 
Wang Mang was in the saddle. At his orders one was built 
there, timed to usher in his new dynasty. On what pattern 
and on what suburban site is not stated in the "History of 
the Former Han Dynasty" (either in ch. 25 or ch. 99), nor 
what function it fulfilled. It disappears from the records after 
Wang Mang's fall. 

We come to the mmg t^ang of a.d. 59 at Loyang. I submit 
that on the above evidence the Former Han government 
never sacrificed to the Five Ti in a mmg fang at Ch'ang-an. 
Since the rites were felt to be vital for seasonal equilibrium, 
Kuang Wu at the end of his reign decided that he must have 
them, and have them on the spot. He desired Emperor Wu's 
numinous power, and the new capital needed to be conse- 
crated, established in men's minds as the main centre of aus- 
picious influences. Kuang Wu died before the buildings were 
completed and Emperor Ming completed them within some 
twelve months. Emperor Wu's cabalistic design was followed, 
but innovations were made in connection with the robes. 

With regard to Kuang Wu as "/)^^i," ancillary to and as- 
sociate with the Five Ti, the "History of the Later Han Dy- 
nasty" (ch. 18) states that there was an altar for him south 
of that to the Green Ti and facing west. He received his 
sacrifice of a bull along with each of the Five Ti. In esti- 

there is a very obvious connection between his making King- Wen a f^ei 
to heaven and Emperor Ming making his father a f^ei to the Five Ti. 
® The records are incomplete and we cannot assume that these sacrifices 
were made yearly. 

[ 168 ] 



PAN KU ON THE EASTERN CAPITAL 

mating what his precise status and function were, reference 
naturally should be made to the first poem of the King Wen 
decade in the "Odes." King Wen is there plainly represented 
as on the one hand having achieved superhuman status, on 
the other hand as not only protecting his descendants but 
also keeping a disciplinary eye on them. Thus the Loyang 
Ming T'ang rites embodied two of the traditional purposes, 
the placating of certain Nature-deities and the exaltation 
of a dead father by his filial son. As for the other function, 
a holding of audience and a publishing of policy, there is 
no convincing evidence of such. The nature of the buildings 
was hardly suitable for such a meeting. It is, however, quite 
clear that high dignitaries were present at the sacrifices and 
took part in the procedure as assistants. Visiting vassal kings 
might be among them, brought in to be impressed by this 
demonstration of Chinese civilization and heavenly prestige. 

The Pi Yung Ode gives no indication that the Pi Yung 
was the centre of higher education, nor does it refer to the 
annual archery event, both features a commonplace to Pan 
Ku's readers. Its concern is confined to the encircling moat 
and the Emperor's entertainment in person of the state vet- 
erans. Quite clearly Pan Ku attached importance to the caba- 
listic significance of that circle of running water. The question 
is whether there was such in the imperial college in Ch'ang-an. 
The evidence is not easy to find, but since archery was part 
of the old-time education for the nobility, there is ground 
for the supposition supported by some scholiasts that for pur- 
poses of public safety the college campus was surrounded b\' 
a moat. As for possible innovations in relation to the Loyang 
emperors' shooting with the bow and their serving of the 
veterans, Chang Heng's Loyang fu is so much more detailed 
that the critique will deal with these matters there. 

For the Loyang ling t^ai it must suffice to emphasize two 
aspects: (i) Pan Ku is of course correct in his emphatic "long 
venerated" in description of a royal observatory. There is the 
''Ling rap' Ode in the "King Wen" decade of the "Odes," 

I 169 ] 



CRITIQUE 

and no question but that the Chou kings regarded the posses- 
sion of an observatory as one of the main royal duties and 
prerogatives. (2) To judge by the monograph on "Music 
and the Calendar" in the "History of the Later Han Dy- 
nasty," Emperor Wu's realignment of the year in 104 b.c. 
came up for criticism in Kuang Wu's reign, and there was a 
good deal of discussion then and later, the redoubtable Chia 
K^uei being one of the disputants. The matters under dis- 
cussion were highly technical and must be taken as being 
beyond the scope of this critique. The frima facie assumption 
is that where there was so much discussion there were new 
ideas brewing in the men's minds. 

In conclusion, on this fu as a whole, the impression grows 
steadily stronger that the change of capital was accompanied 
by a number of striking innovations. These changes were 
not only in relation to institutions but also to certain more 
or less definable movements in the mind of the age. On that 
basis of consideration it is worthwhile trying to define these 
movements here (this in preparation for more systematic con- 
sideration in Chapter XI). On the one hand, the govern- 
mental machine in Loyang was more clearly, and even 
designedly, constructed on a constitutional basis than the 
Ch'ang-an machine ever was — a development marked, for 
example, by the establishment and rituals of the "Three 
Yung" (Ming T'ang, Ling T^ai, and Pi Yung). On the 
other hand, the adaptation of ancient procedures to the newly- 
envisaged requirements of government was carried out in a 
spirit which was as superstitious as it was intelligent. The 
tug of war in Pan Ku's mind between an animistic logic and 
a more rational natural-cause-and-effect logic has from time 
to time emerged very clearly to view. In the third place, 
in regard to "Confucianism" (so called), those enhanced and 
rationalized values attached to the worship of the Five Ti 
would seem to have little connection with the influence of 
Confucius, the sixth-century-B.c. sage of Lu. Yet the sheer 

[ 170 ] 



PAN KU ON THE EASTERN CAPITAL 

prestige of his name, indefinable yet powerful, becomes in- 
creasingly discernible in the last seven reigns of Former Han. 
Something of immense doctrinal significance was coming 
through to consciousness, something very much alive but as 
yet inchoate, something which found clearer expression after 
the scholars' experience of the Wang Mang episode. One 
thing is clear. Although the consciousness of world-empire, 
its glories and duties, brought the advance to a universal 
ethic, yet neither in its earlier phases nor in its Later Han 
developments was that ethic the concomitant of a monotheis- 
tic religious consciousness. 



[ 171 ] 



Chapter X 



CRITIQUE OF CHANG HENG'S 
"FC7 ON THE EASTERN CAPITAL" 

I. The Teacher^ s Reaction 

Whereas Pan Ku made his Loyang host deal in very super- 
cilious, not to say snobbish and derisive, fashion with his 
provincial guest, Chang Heng took a strikingly different 
line. His old teacher is represented as disconcerted by the 
nobleman's cynical materialism. However, he pulls himself 
together and is able to start his rejoinder with a smile. Un- 
questionably, Chang Heng's dramatic psychology is an im- 
provement on Pan Ku's crudity. Also, it shows a better ap- 
preciation of good manners, as we would expect from a mem- 
ber of that highly-mannered Loyang court. Here is the man 
of breeding {chun-tzu) in action, as the student has come to 
recognize him in the Analects: the gentleman who does not 
ride roughshod over his opponent's feelings. That being so, 
we come to see that in that court both dramatizations could 
be enjoyed, the somewhat vulgar as well as the more pol- 
ished : in the latter case the teacher's opening words are blunt 
enough in spite of his smile. There is, for example, the sar- 
castic comparison with the poor outlawed minister among the 
barbarians, Yu Yu, of the seventh century b.c. How could 
it be that he could see things in their right perspective, whilst 
"you, my dear sir, with all your knowledge of past and pres- 
ent, could be as misguided as you are?" 

The nobleman is roundly accused of "prizing the ear," i.e. 
hearsay, and "despising the eye," i.e. ocular demonstration. 
In other words, under the urge of masterful desires, he is 
guilty of wilful self-deception, and able to prove anything 
to his own satisfaction. Since in his final remarks on the 
Ch^ang-an period he had wound up with the complaint that 

[ 172 ] 



CHANG HENG ON THE EASTERN CAPITAL 

an aristocrat's taste for elegant luxury was nowadays stupidly 
restricted, the charge was true enough. Lord Chesterfield, 
when taken to court by his pastry-cook for not paying his bill, 
remarked to the judge that it was a bit hard if a gentleman 
could not be allowed to enjoy a biscuit with his sherry j and 
it was this combination of frivolity and class egoism which a 
generation later brought England to the verge of bloody rev- 
olution. In Chang Heng's eyes, for all the virtues of the new 
regime, this attitude of mind was still alive and a grave dan- 
ger to the common weal. The couplet dealing with the hy- 
pothesis of a breast {hsiung) without a mind {hsin) deserves 
by its luminous simplicity to be made a locus classicus. The 
trouble about Han psychology for the modern reader is its 
confusing complexity, as may be seen in the pages of the 
Po Hu T^ung, notably those on the human frame, including 
hsin in particular. Presumably Chang Heng had very little 
use for that kind of pseudo-logical synthesizing. It was the 
last generations of the Chou era which saw the rise of per- 
sistent introspective analysis, witness Mencius, Hsiin Tzu 
and the Lil Shih Ch^un ChHu^ not to speak of Chuang 
Chou's metaphysico-psychological theorizing. It is, therefore, 
against that complicated background with its wealth of terms, 
all of them used with varying connotations, that we come 
to appreciate Chang Heng's blunt extrovert approach with 
its hsiung as the seat of the emotions, passions, appetites, and 
its hsin as that directive intellectual force in man which can 
control those same unruly emotions.^ Also, the Western stu- 

'^ ^^Hsin,^^ one of the comparatively few clearly recog-nizable pictographs 
in its pre-Ch'in form, does not appear till a Chou II inscription: so B. 
Karlgren's Grammaia Serica states. In the Chou Odes it is found constantly, 
and, as one might expect in that stage of cultural development, denotes the 
physical heart as the seat of both feelings and thoughts. Hsiung occurs once 
in Menc'ms (IV A 15) and twice in the Wen Sang of the "Record of Rites" 
and in the Lu Shih Ch^un ChHu. In the Mencius instance it has a vaguely 
psychological connotation; in the other instances it denotes the physical 
breast. By Chin times it came to be associated with thoughts as well as 
feelings. There are no instances that I can find of thoughts being associated 
with the head, i.e. the brain. Hsin^ on the other hand, was used right along 
in relation sometimes to feelings, sometimes to thoughts. Yet by late Chou 

[ 173 ] 



CRITIQUE 

dent needs to appreciate Chang Heng's immediate associa- 
tion of this intellectual force with clarity of moral vision. In 
this respect, although Taoist speculations threw doubt on 
the ultimate cogency of morals, the dominant tendency in 
the scholars' thinking was to make that direct association. 
They linked reason with a categorical imperative which func- 
tioned by means of the accepted code of morals and good 
manners. The same tendency prevailed in classical European 
thinking from Plato down to the Roman Stoics. 

With regard to the consciousness of hearsay as unreliable, 
as we have seen. Pan Ku's historical studies had brought him 
to the same position. There was Wang Cheung also with his 
pejorative attitude. It would, however, be a mistake to re- 
gard this critical outlook as new to Later Han. As far back 
as the last days of the Chou era there was discriminate think- 
ing on this score. Thus in the Lii Shih Ch^un Ch'iu we find 
(xxii, 6) the following statements under the caption of 
"Examine Tradition": "Traditional statements should not 
fail to be examined. There are some traditions in which 
white becomes black. . . . To hear and examine is happiness: 
to hear and not to examine — it were better not to hear." 
The author then gives some simple illustrations from tradi- 
tional history which prove his point. What may well be new 
in Chang Heng's outlook here is his association of this un- 
reliability of hearsay with the complexity of past history, his 
moral being that the ethical issues are by no means easy to 
elucidate. He proceeds immediately to exemplify this. 

2. The Real Lesson of the Dynastic Revolutions 
{Third Century B,C.) 

As the opening section of Chang Heng's ^^Fu on the West- 
ern Capital" shows, the nobleman's approach to history is a 

times, there is no question, thoughts and feelings were sharply distinguished, 
and a further complication arose by the use of chih^ originally denoting a 
concrete aim, coming to be used in the sense in which a faculty psychologist 
would use it, namely, denoting the will. At the same time the practice grew 
of regarding the hsin as the lawful lord of the individual, controlling de- 
sire. Obviously, that was Chang Heng's position in his use of hsin here. 

[ 174 1 



CHANG HENG ON THE EASTERN CAPITAL 

cynically amoralistic one. Eastern Chou, being centred in a 
poor-soiled area, was therefore weak and could not stand up 
against the powerful Ch^n with a rich-soiled area in its pos- 
session j and whatever royal line rules, if they are rich, they 
are sure to live luxuriously — indeed, why should they not? 
Now comes the Teacher's reply. Chou in the end was in- 
capable of governing and had to die, but each of the Seven 
Powers was disqualified by its immoral appetite for display. 
The First Emperor, like a tiger, like a bird of prey, could 
not be withstood. But when all belonged to him, he wanted 
all the enjoyment for himself, working his people to death 
on his megalomaniac projects. 

The Teacher's theme, indeed Chang Heng's theme, is the 
one familiar to Western historians, the Greek sense of a 
nemesis awaiting the ruler guilty of hubris. But here the 
chain of cause and effect operates on a wider scale. The doom 
is not the inescapable one operating among the members of 
a royal family, so that the best of them cannot escape de- 
struction, but the doom of a cruel oppressor of the people. 
Chang Heng was a central China man, and he conveys with 
a rugged poetic eloquence the horror the central China people 
had felt when that half-barbarian tyrant from the west drove 
them worse than their own deboshed traditional rulers. And 
in the chain of events there is no sign of this and that god 
on Olympus pulling the strings. There is just a bare refer- 
ence to Kao Tsu as "obeying Heaven." The doom of Ch4n 
came from the people: the "Hundred Surnames" (note the 
suggestion of family affiliation) "could not endure the usage" 
they received. And Kao Tsu's power of sovereignty lay not 
only in mighty deeds of war but in his plain man's revulsion 
against unnecessary pomp and luxury. 

Those "western craftsmen" deserve attention. Emphasis 
has already been laid on the stimulation of skilled craftsmen 
resulting from the imperious demands of wealthy courts. 
Here is another illustration, pinpointing the fact at the earlier 
date of the third century b.c. Thus the Ch^in First Emperor 
collected a band of the best artisans, so that what had oper- 

[ 175 ] 



CRITIQUE 

ated on a small scale in the Chou state courts operated on a 
considerably increased scale in the Ch^in capital of Hsien- 
yang. These men became a permanent part of the court 
establishment, and, since expense was no bar, it paid to stir 
the more able craftsmen by rewards. Whether Chang Heng 
had actual information on these Ch4n craftsmen and "their 
eyes playing over O Pang" we do not know. From the psy- 
chological point of view it was the natural thing for them 
to do, and this was part of Chang Heng's sympathy with 
workers and their workmanship, to envisage the situation 
and their disappointment when they found their new master 
scornful of their efforts. 

To take a wider perspective, there is no question about the 
lasting dint which the Ch'in tyranny made on Chinese con- 
sciousness in the Han era. In this connection Pan Ku's empha- 
sis on Kao Tsu being the first plebeian to win a throne proved 
highly suggestive. Chang Heng for his part had nothing to 
say on that subject, but the same feeling is discernible in him, 
viz. that the coming of Han was an event of unprecedented 
significance. Viewed from the later angle of vision the old 
ruling families had been hopeless, the new Ch4n a nightmare. 
Then came this plebeian with his indomitable energy, his 
resourcefulness in planning, his downright scorn of court 
flummery : the despairing people could rest their aching shoul- 
ders on him. Historians today do well in recognizing his 
limitations of outlook and in discounting Chia Ps (199-168 
B.C.) highly-colored account of the peasants everywhere 
marching to his support, but the question stands whether that 
typical aristocrat, Hsiang Yii, and any body of traditionally- 
minded Lu scholars could have created a new will to live, 
and live under one government. The old regional semi-tribal 
jealousies were still strong and were to remain so. Yet with- 
out the impetus given by the man from P^ei there might so 
easily have been no united China, indeed no China of the 
Sages, with its notable cultural achievements. The First Em- 
peror with his forty commanderies had roughed out a frame- 

[ 176 ] 



CHANG HENG ON THE EASTERN CAPITAL 

work of imperial dominion, but his regime flouted the needs 
of the "Hundred Surnames" and was bound to collapse. So 
Pan Ku and Chang Heng saw it. With Kao Tsu began an 
unprecedented type of unification, one in which for all its 
shortcomings the theories of the Chou philosophers slowly 
took shape in a constitution under which the common people 
were recognized as having rights. 

3. The Crossness of the Nobleman* s Blindness 

The prosody is irregular, with no concern for rhyme. In- 
deed, the poem here has more the swing of dialectical prose, 
as the Teacher sets forth the moral perversity of the Noble- 
man's position. That position is shown to be not merely per- 
verse, but also rationally absurd, since it made nonsense of 
vital events in the history of the sage-kings. The Nobleman, 
then, is denounced as "not knowing the force of words" {fu 
chih yenY — a very sharp accusation to bring against a cul- 
tured gentleman, and so revealing the strength of Chang 
Heng's feeling. 

With regard to the preservation of the spirit tablets, the 
main significance of the reference is clear: all the earlier Han 
emperors down to the end, whatever the vices of the later 
ones, had faithfully done their duty by their ancestral tab- 
lets. The Teacher is not out to whitewash the evils attend- 
ing those later reigns, but in his eyes the plain fact that the 
ancestral fanes were kept in repair and the order of sacri- 
ficing maintained was proof of effectual filial piety, and such 
filial piety fer se was virtue of a high order. There can be 
no doubt that the Late Chou and the Han eras produced a 
hs'mo tao ("way of filial piety") which in its ethical refine- 
ments as in its ritual forms went far beyond the original 
ancestor worship with all its fears of hungry spirits bringing 
calamity on their descendants. Nevertheless the old fears re- 
mained, as is plainly seen in the struggle over the problem 

^ An allusion to Analects^ XX, 3: "He Avho does not understand words 
cannot understand people." 

[ 177 ] 



CRITIQUE 

of the funerary parks between 40 and 33 b.c. (see Dubs, II, 
289-291). 

Again the historian is faced with an amorphous, fluid ideo- 
logical situation, in relation to which sweeping generalizations 
are bound to create misleading impressions. For instance, how 
are we to appraise Kung Yli and Wei Hsuan-ch^eng in their 
rugged common-sense insistence that many of the funerary 
parks were redundant and should be abolished? Were they 
being rational, liberal, conservative, superstitious? Read their 
memorials, and it is clear that they were all four. And what 
motive lay behind the innovation of having in addition to 
the High Founder's fane three other comparable fanes dedi- 
cated to the great Tsung ("Exemplars"), Emperors Wen, 
Wu, and Hsiian? Was that done because in life they had 
been such effective individuals that their inspiring influence 
was, rationally enough, mobilized to the support of the 
throne? Or was it just fear of their ghostly mana? Further, 
for all its sentimental idealism, the Han hsiao tao was realistic 
enough in a callous sense. After five generations the tablets 
of the sacred dead were normally moved into a side room, 
and there in course of time they became mere decaying 
lumber. 

4. The Rightness of Loyang as the Cafital 
for True Sons of Heaven 

The poem, getting into its metrical stride with solid rhyme 
patterns, comes to its first main theme, the rightness of the 
Loyang site. As might be expected, Chang Heng details the 
natural amenities of the district and attaches importance to 
the fact that in the days of old this was made a centre of 
government. These features are, however, only cited as con- 
firmatory evidence. The basic argument admits by implication 
that the new capital is less well-guarded by mountain defences 
than the old capital, but lays down that in the last resort 
the throne is protected by the good faith {Hang) of the ruler's 
subjects. The theme, speaking generally, has a familiar ring 

[ 178 ] 



CHANG HENG ON THE EASTERN CAPITAL 

to the student of the Analects and Mencius, and the reader 
may be tempted to take it for granted that Chang Heng here 
enunciates what in principle had been accepted by the scholars. 
There is room for doubt, since the Teacher is made to draw 
a sharp antithesis between reliance on passes and reliance on 
the people's good faith. Pan Ku's Loyang /«, for all its elo- 
quence about popular gratitude over peace and plenty, has 
nothing so explicit as this, and we know from the "Acts of 
the Emperors" that the Loyang government made full use 
of the military arm, both for protection and imperial expan- 
sion. The general idea was that both jen and military effi- 
ciency were a sine qua non of successful government. That 
was so in Earlier Han times, and the question about which 
was more important could hardly arise until the establish- 
ment of a less naturally protected capital. 

Ancient Chou is presented as the discoverer of the site: 
King Ch'eng and his faithful servants, the Dukes of Chou 
and Shao. Although the two Dukes are not mentioned by 
name until later in the section, they are the master figures 
in the story, particularly Duke Chouj and the language of 
the poem where it treats of them is reminiscent of the Shao 
Kao and Lo Kao sections of the "History" Scripture where 
the two Dukes occupy the real centre of the stage. There is, 
however, in those documents no mention of a t^u kuei (meas- 
uring table) and calculations being made: they refer only 
to divining and finding the auspices good. For mention of 
a t'u kuei we have to go to the Chou Li, e.g. Ti Kuan Ssu T^u 
(ch. 2): "[The officer], using the method of the measuring 
table, calculates the relative depth of the earth,^ verifying 
the shadow of the sun with a view to finding the central 
point of the land surface."* From this and other passages 

^ "Relative depth" here refers to the depth of the shadow cast on the 
t^u kuei plate. The reference can only be to the relative proximity of any 
locality to the four points north, south, east, and west. 

*The passage continues: "[In the case of] the sun being- southerly [in 
relation to the t^u kuei], its shadow is too short and [the locality] has too 
much heat. [In the case of] the sun being northerly, its shadow is too long 

[ 179 ] 



CRITIQUE 

in the Chou Lr' it is clear that the t^u kuei was an oblong 
plate on which fifteen inches (Chinese) were marked in re- 
lation to a perpendicular pole of eight feet (Chinese) high. 
The history of the experimentation from which this table 
of measurement came to completion is entirely obscure, ex- 
cept for the probability that in Eastern Chou times there was 
practical need for measuring the boundaries of the lOOO // 
royal demesne (cf. Chou Li^ loc.cit.). In any case, the dis- 
covery was made, so it was thought, that where a horizontal 
plate placed due north and south with a perpendicular pole 
eight feet high at its southern end threw a fifteen-inch 
shadow, there was the centre of "the land." It is not clear 
whether the rudimentary experimenters took adequate steps 
to ensure that the pole was exactly at right angles, although 
they became alive to the necessity of achieving rigidity of 
position in the pole.® Nor is it clear whether they took ade- 
quate steps for ascertaining exact horizontality for the plate, 
e.g. by judging from the surface of a pond. 

Here, then, is unmistakable evidence that the genuine 

and [the locality] has too much cold. [In the case of] the sun being easterly 
[? westerly], its shadow has a post-meridional [slant] and [the locality] 
has too much wind. [In the case of] the sun being westerly [? easterly] its 
shadow has a premeridional [slant], and [the locality] has too many rain 
clouds." The Chinese only makes sense, if the author is assumed to be 
thinking of (i) a standard-ruled t^u kuei being taken round to four lo- 
calities assumed to be south or north, etc. 5 (2) the operators having with 
them a clepsydra which registered midday for the summer solstice at the 
central point. The scholiasts support this interpretation, but they seem to 
differ in opinion as to whether ^^jih tung'^ (the sun being easterly) refers 
to an eastern or a western locality. To me these words do not make sense 
unless the reference there is to a western locality. The sun in a western 
locality could not but rise later than in an eastern, and given these ancient 
Chinese observers had a center-standardized clepsydra, they would see from 
the post-meridional slant of the shadow on the plate that the sun had not 
yet reached its meridian. That being so, the experiment must have been in 
a western locality not an eastern. By parity of reasoning the fourth alterna- 
tive observation introduced by '•'■jih hsV (the sun being westerly) must have 
been made in an eastern locality, not a western. 

° See H. Maspero, Les htstruments A stronomiques des Chinois au Temfs 
des Hariy Melanges Chinois et Bouddhiques, VI, Bruges, 1939, p. 220. 

® See Maspero, of.cii.y pp. 218-219 and Tung Tso-pin, C/iou Kung Ts'e 
Ying T'ai Tiao-ch^a Pao-kao^ Academia Sinica Bulletin, 1939. 

[ 180 ] 



CHANG HENG ON THE EASTERN CAPITAL 

scientific mind was at work, making experiments in an effort 
to establish earth-space on a mathematical basis of accuracy. 
The scientific urge in this is the more impressive because — 
at any rate by Han times — it was realized that the only 
feasible method was to equate space to time, namely by as- 
certaining from the movements of the heavenly bodies at 
what times the winter and summer solstices came in relation 
to the capital. Thus the invention of the sundial and the 
clepsydra were stepping-stones in the quest for the accuracy 
of dial readings. That those would-be scientists were wrong 
in the assumption that the earth was flat, not globular, was 
their mistake, as also their assumption that the area of their 
national habitat contained the centre of the terrestrial world. 
Thus they were ignorant of the fact that their country lay 
north of an equatorial global division and was only part of 
one of two longitudinal hemispheres. But this ignorance of 
theirs does not detract from the scientific aim of their experi- 
mentation, nor from the validity of their discovery that space 
and time can only be calculated on the basis of their being 
correlatives.*^ 

In the light of the above, what, then, is the evidential 
value of Chang Heng's reference to the /w kuei in connec- 
tion with the establishment of the Loyang capital? First, it 
is significant that, although the Shao Kao and Lo Kao records 
of the first establishment of Loyang have no suggestion that 
any geodetical experiments were made, yet Chang Heng as- 
sumed that they were made. Since the Scriptures were en- 
tirely precise in attaching the fact of that establishment to 

"^ Maspero (op.cii.) had a good deal to say on the final accuracy of the 
calculations in respect to the i'u kueiy and on p. 223 he questioned whether 
the Chinese constructors of these t'u kuei which have survived were so 
anxious for exactitude. The justice of his critical reflection in those con- 
nections is not to be questioned, but I find a particular interest in the fact 
that archaeological evidence of a t^u kuei is obtainable from, for example, 
the back-blocks of the empire. It points to the extensive use of the instru- 
ment and opens the door for the conjecture that some t^u kuei were made 
under insufficient expert supervision. It does not seem to me that the care- 
lessness thus evidenced can be assumed as applying to headquarter experts 
of the scientific calibre of a Chang Heng. 

[ 181 ] 



CRITIQUE 

King Ch'eng and his famous ministers, here is evidence that 
around a.d. i 20 one highbrow was not satisfied with divinatory 
readings as sufficient proof of the rightness of the site. In 
his judgement there must have been t^u kuei measurements. 
Also, in his eyes the importance of getting the right site 
derived from climatic considerations, the avoidance of ex- 
cessive heat and cold, wind and rain. 

Second, there seems no adequate ground for doubting that 

in the early days of the Chou regime, a royal administrative 

centre was established at Loyangj and certainly this was 

later the site of the capital for the East Chou regime. The 

climatic amenities must in due course have become clear from 

ordinary year-to-year observation. Yet in Han times this idea 

I of t^u kuei observations being a necessary preliminary to the 

j- discovery of the site came to be accepted. How could that 

j have happened? There seems only one way, and that the 

f one illustrated in the Hsi Tz'u ("Appended Amplifications") 

!. to the "Changes" Scripture: the passage in the second part, 

' where the main implements conducive to man's well-being 

? are attributed to discoveries by the sage-kings, and to them 

i being guided by the logic which, it was now maintained, was 

L latent in the diviners' hexagrams. The Han attribution of 

the use of the fu kuei to Duke Chou is on a par with this 

kind of reasoning: the decision for Loyang was a sage-like 

decision, and Duke Chou was a sage, and therefore he used 

a sage-wise instrument. 

Here, then, lies the evidential value of Chang Heng's 
statement. It brings us near to the new urge towards scientific 
certainty which came in the mid-Han and Later Han periods. 
To the questing minds of that epoch the sage-kings of an- 
tiquity were not merely mystically inspired by Heaven, but 
even more intellectually stimulated to explore the rational 
elements in the physical universe. For further illustrations 
(apart from those in this book) of this new attitude of mind 
see the "History of the Later Han Dynasty," ch. 12, e.g. in 
the thesis there propounded by Pan Ku's contemporary, Chia 

[ 182 ] 



CHANG HENG ON THE EASTERN CAPITAL 

K^uei, on the rectification of the calendar. It is that although 
astrological thinking by no means disappeared there was nev- 
ertheless a new class of scientifically-minded astronomers. 
Chang Heng, it may now safely be inferred, was an out- 
standing example of this class of men whose conclusions were 
derived from the use of the clepsydra and the armillary 
sphere (hsuanchi). 

With regard to Chang Heng's denunciation of Wang 
Mang, it is less detailed than Pan Ku's but along much the 
same line: vilification of a man who tried unsuccessfully to 
found a new dynasty. On the other hand, there is a very 
revealing expression: "throughout the nation no one dared 
have another opinion than his." In the light of what is known 
about Wang Mang's career, it can hardly be questioned that 
he was highly dictatorial in temperament. 

5. The New Palaces and Pleasaunces 
Described and Aff raised 

In these forty-two couplets the narrative moves from item 
to item, the transition marked by change of rhyme and pro- 
sodic rhythm. From the point of view of orderly description 
this section is a masterpiece, true to the genius of double- 
harness composition. But the narrative is so compressed that 
there is nothing to show whether any of the buildings speci- 
fied were administrative offices. Presumably some were, for 
the Ming T^ang, Pi Yung and Ling T^ai (the Three Yung) 
were not places where the day-to-day administration was done. 
Chang Heng's treatment stands in sharp contrast to Pan 
Ku's, for the latter gives no detail whatever, speeding along 
to his great emphasis, the noble frugality which inspired 
Emperors Kuang Wu and Ming. Chang Heng also praises 
the moderation shown, but, as one high-sounding appellation 
succeeds another, the seed of a suspicion is sown. One wonders 
whether a tincture of delicate irony does not permeate the 
account as a whole. 

Finally there is "the laborers labored, the dwellers are 

[ 183 ] 



CRITIQUE 

at ease." The T^ang commentator, Chang Hsien, takes these 
words as conveying the emperor's attitude. Assuming this 
conjecture to be right, we have to face Li Shan's comment. 
That cites a story from Chia I's Hsin Shu^ in which a cer- 
tain king of Ti sent an ambassador to the King of Ch^u. 
The ambassador, being entertained in a very ornate building, 
was asked whether his master had any to compare with it. 
He replied, "The thatch has not been trimmed, the beams 
not cut to length. Yet my king still feels that the builders 
have been too hardworked and that the dwellers are too 
much at ease." The emperor referred to in Chang Heng's 
passage was Emperor Ming, for he is the central figure 
throughout the section. That being so, we are faced with 
the fact that Chang Heng clearly had a genuine admiration 
for him. Further, the passage in question comes next to the 
building of the Three Yung. There is no evidence that they 
were ornate buildings j indeed, it is quite to the contrary. 
Thus the final conclusion is that our author may well have 
inserted a delicate probe or two, but it is doubtful whether 
he wrote in the ironic vein. This conclusion releases our 
minds to appreciate to the full that this section is no dreary 
sycophantic recital of the beauties of Loyang. On the con- 
trary the poet was in command of his material, so much so 
that his wit could flicker into action. 

In the main the general appraisal corroborates Pan Ku's, 
and, as the critique has shown, there is good ground for ac- 
cepting it as approximately true to the facts. Pan Ku was 
lyrical about the new architecture, calling it "divine art." 
Chang Heng more soberly sums it up as "economy not re- 
sulting in crudity [/o^]." Since the two men saw the build- 
ings with their own eyes, it would be of great interest if 
the archaeologists of today could find evidence pointing to 
this new motif of restraint. 

The mind of our poet-scientist is a study, as it moves from 
point to point of appraisal. From restraint not producing a 

8 Chia I, Hsin Shu, ch. 7, "T'tt/ Jang:' 

[ 184 ] 



CHANG HENG ON THE EASTERN CAPITAL 

debasement of art, that mind moved — may one say natu- 
rally — to the observance of old and simpler standards, and 
a set goal to be achieved. (Scientists seem to have a proclivity 
for that kind of thing, definition of the goal and precision 
in the means for attaining it.) Then comes "Behold, here 
is Lij the right and fitting, in full display." Yes, of course! 
That was it, //, the uplifting of a standard of the spirit, the 
ethically right and aesthetically fine embodied in a set of 
harmonious actions. And that range of ideas brought to 
Chang Heng's mind King Wen, of sainted memory, build- 
ing his Spirit Terrace ("Odes," III, i, 8): no driving of 
the workmen, their task accomplished with a will. That also 
was //, the demonstration of fellow-feeling between master 
and man, between class and class. It crops up again at the 
end of the section, and, indeed, is a major theme of the poem, 
recurring again and again. 

With regard to the Ming T'ang, Pi Yung, and Ling T'ai, 
the order of narration makes them the climax of the building 
operations. Yet there is no sign of the lyrical fervor Pan 
Ku felt about them. In plain four-word sentences the bare 
essential facts about their structure are stated and their re- 
spective functions outlined. Emphasis is laid on the elaborate 
symbolism of the Ming T'ang, its set-up being along the 
lines of Emperor Wu's ming fang. That sacrificial rites and 
recruitment for state service and observation of weather 
should all come under one generalization needs no comment; 
but the emphasis on principles does. The yearly routine of 
the Three Yung was a public demonstration of government 
at the highest level being by unchanging rules of procedure. 
That looks suspiciously like recognition of constitutional prac- 
tice. The Pi Yung rewards scrutiny. The stress is laid on 
it as the place where the selection of candidates for office 
took place. In that respect examinational procedure in the 
Imperial Academy becomes complementary to the develop- 
ment of constitutional government. If the "Record of Edu- 
cation" {Hsueh Chi) in the "Record of Rites" may be taken 

[ 185 ] 



CRITIQUE 

as having a close relation in time to Chang Heng's age, we 
can envisage Chang Heng as thinking of those yearly college 
tests being carried out, and the final test coming after nine 
years of study. It should be borne in mind that the Pi Yung 
site included archery butts and proficiency in archery was 
one of the tests — proficiency in marksmanship as also in cour- 
tesy under the strain of competition. To go by the "Signifi- 
cance of Archery" {She I) in the "Record of Rites," there 
was more attention paid to physical stamina than has perhaps 
been recognized so far in Sinological studies. 

The treatment of the Ling T'ai makes one rub one's eyes. 
Here is Chang Heng speaking of his own special field of 
study, and all he has to say is that the P4ng Hsiang ( official) 
makes observations of evil Yin- Yang weather elements at 
strife and has intercession made for their removal. This of- 
ficer's duty seems to have been to keep an eye on the stra- 
tegic points in the seasons and give warning of threatening 
changes.^ Warning being given, the right prayer formula 
might be used to avert them. Now, why did Chang Heng 
treat the Ling T'ai theme like that? Was he in revolt against 
Pan Ku's rosy picture in the ode? Or did the decline in court 
morale in Emperor An's reign bring the exaltation of minor 
officials like the P^ng Hsiang and so depreciate the services 
of the more expert astronomers? The answers to these ques- 
tions can only be conjectures. 

6. The Einfire and Beyond Doing Homage 

Again there is a magnificent piece of vivid description, the 
swing of it from point to point marked by rhyme and rhythm 
changes. At certain key points the prosody is irregular — very 
skilfully so since the irregular lines make a pattern of em- 
phases. For example, the crowning point of the proceedings 
when the vassals present their tribute and the Son of Heaven 

^ For this title, see Chou Li, ChUm Kuan, Tsung Po, where ^^P'ing Hsiang^'' 
is mentioned after '•'■T'ai ShW (the Grand Astrologer). No such title 
occurs in the monographs on the Hundred Officials and on the Calendar 
in the "History of the Later Han Dynasty." 

[ i86 ] 



CHANG HENG ON THE EASTERN CAPITAL 

responds with the properly graded bow. This is marked by 
an eight-word line followed by a ten-word line, the latter 
not fitting into a rhyme scheme. 

There is vivid delineation of the scene, as if coming from 
an eye-witness. Yet the special durbar on a grand scale with 
a large gathering of foreign princes held by Emperor Chang 
was in a.d. 87 j and since our author was only nine years old 
then it seems doubtful whether he saw it with his own eyes. 
There were, however, other durbars, for in theory and more 
or less in practice, the homage of the vassals was due every 
New Year's day. (Travelling in winter must have been a 
trying business for those from distant parts in the north.) 
I assume that Chang Heng's picture was a conglomerate one. 

There is not a single word to indicate where the audience 
was held. Fortunately Pan Ku's account refers to a "Cloud- 
Dragon Court" and Chang Heng in Section 5 has specified 
a Cloud-Dragon Gate to the palace enclosure. It was on the 
east side, corresponding to a complementary Spirit-Tiger Gate 
on the west side of the enclosure. Since the Emperor on his 
throne faced south, these two gates presumably lay south- 
east and south-west of the great audience hall. The difficulty 
is to envisage space within one courtyard, however big, for 
"several tens of thousands" of onlookers.^^ Presumably there 
is considerable exaggeration. However, the account accurately 
distinguishes between the onlookers and the actual partici- 
pants in the audience: princes of the blood, foreign poten- 
tates, enfiefed nobles, and high officials. They along with 
the larger crowd are divided into two sections, presumably 
making a lane for the imperial carriage. We can imagine the 

^^ These onlookers should not be taken to be the common people but 
minor court officers, members of the princes' retinue, anyone who had 
entree to the palace enclosure. Pan Ku referred to tents being erected in 
the Cloud-Dragon courtyard, and that may be taken as being so on the 
occasion of Emperor Ming's smaller durbar. There is no need to assume 
that for Emperor Chang's great durbar and subsequent ones there were 
tents for the barbarian guests inside the palace enclosure, or at any rate 
in the Cloud-Dragon courtyard. 

[ 187 ] 



CRITIQUE 

hush as it passed through the throng — broken by the crash 
of the sound from the drums and bells. 

As a piece of pageantry it rivets attention, described, it 
would seem, from the viewpoint of someone in the main 
crowd. First the serried ranks of dignitaries, then the bells 
and drums on their stands, the guards with crossed halberds 
on the steps to the hall, the imperial carriage filling the 
porch, the banners a-flutter! Then the torches a-blaze in the 
audience hall and the crash of the salute with the tremor 
of the earth! There is the summons to stand apart, as the 
Emperor alights and enters the robing room. Majesty emerges 
in full regalia and takes its seat on the throne. Then comes 
the great moment, the Emperor faces due south, and the 
if audience has begun. 

'} There seem to have been two flights of steps, one where 

:i the guards stand, outside the hall, the other inside. Accord- 

^r ing to the commentator Hsieh Tsung, this latter consisted 

J- of three parallel flights leading up to the throne, the central 

• flight used by the Emperor, the eastern flight by the inner 

\\ ring of dignitaries when invited to confer. The odd thing 

•|:^ is that no mention is made of the vassals doing obeisance, 

\ only of the "hundred princes" entering the hall and there 

a- being marshalled in order of rank by the ushers. Was this 

I omission due to Chang Heng, from where he stood, being 

3 unable to see the actual kow-tows? He could see the tribute 

a treasures being presented and the Son of Heaven's graded 

response. 

The comparison of this account with Pan Ku's is "illumi- 
nating. In the first place, although both Chang Heng and 
his predecessor take the all-subduing power of the imperial 
virtus as their central theme, their treatment of it is almost 
totally different. Pan Ku was just lyrically enthusiastic about 
it: "no one on land or water but breathless and trembling 
comes to pay homage." Chang Heng depicts Majesty on 
its awe-inspiring throne, and then is content with "How 
stately, how majestic, [and'] a spectacle to engage the trust 

[ i88 ] 



CHANG HENG ON THE EASTERN CAPITAL 

of all the world!" Having struck that note, he proceeds to 
elaborate: the Son of Heaven calling his high vassals and 
responsible officers to confer on the "ten thousand chP^ (lit. 
"spring-releases") the endless sudden vicissitudes which at- 
tend the exercise of government. From that he proceeds to 
reflect compassionately on the mass of men, and in particular 
the individual who "has not found his place," i.e. is without 
a family or any economic niche in society. There were many 
such displaced persons in Han China, as repeated edicts re- 
ferring to them show, and in Central Asia a number of dis- 
placed tribes. Right along, the stress is on bearing the burden 
of responsibility. And this linked itself in the poet's mind 
with that distribution of largesse among nobles and officials 
which was a prominent feature of these New Year cere- 
monies. For the Emperor the means to do this, and increase 
official allowances, were there, not so much in treasure stores 
as in the public granaries where the real national wealth was 
held in reserve. Then almost as an afterthought the poem 
comes to the feasting about which Pan Ku had been so elo- 
quent. It disposes of that in two couplets j but concludes like 
Pan Ku with the sense of unity stirred to enthusiasm by 
good food and good wine. 

The poem here has nothing on the music at the feast, 
the national traditional music and the barbarian music, the 
theme with which Pan Ku had so brilliantly reinforced his 
rhapsody on unanimity. In its place Chang Heng pictures 
the Emperor's subjects dispersing happily with a sense of 
duty done, whilst their sovereign lord bore the burden of 
continual watchfulness "and unceasing activity." This is sig- 
nificant if only because it seems so clearly to come by revul- 
sion from Pan Ku's picture of the virtus spreading so natu- 
rally and sweetly among the nations. But there is more to 
it than that. All through Chang Heng's youth and early 
manhood Emperors Ho and An had been the reigning sov- 
ereigns, and they, on gaining the reins of power, showed 
themselves both dissipated and irresponsible. Again we sense 

[ 189 ] 






CRITIQUE 

in our author the purpose of probing the imperial conscience. 
Yet Pan Ku's sense of a natural victory for the virtus is not 
altogether discarded. To Chang Heng it suggested the hsUan 
te ("mysterious, numinous virtus^^) and tzu jan ("self-so- 
ness") of Great Nature in the Taoist view of life, and he 
placed the argument on that transcendent level. Neverthe- 
less, he was no starry-eyed mystic. On goes the probing 
argument to self-examination and to the need for the state 
to get down to grass-roots and seek out "village incorrupti- 
bles" for renewing the highway of the Sages. Only then will 
"high and low communicate the fellow-feeling, and peace 
and joy be here." 

7. The Sacrifice to Heaven 

^L On the first auspicious day in the New Year came the 

S Sacrifice to Heaven, second in the series of ceremonial acts 

sp by which the Emperor kept the ship of state on an even keel. 

'^^ The first five couplets emphasize the reverential care and 

^\ integrity of heart and mind with which the sacrifice should 

J? be approached, the last three couplets the central acts of 

|[ sacrifice, these being preceded by three couplets noting the 

gr elaborate accompaniment of music and ritual dancing. In 

8' between come twenty-five couplets devoted to the pomp and 

I* circumstance with which the occasion was celebrated. 

I First, it is almost impossible to doubt that Chang Heng 

> saw the proceedings with his own eyes, and that he saw them 

'■' from the vantage point of the city gate-tower. A whole suc- 

cession of touches reveal this, e.g. the release inside the city 
of the rows of carriages at the word of command, the color 
schemes of the banners, the tossing of draperies on the cavalry 
horses. From the same vantage point he would hear the 
drums reverberating on the sacrificial site, glimpse the rows 
of dancers, and see the flames and smoke rising, but would 
not descry the Emperor and attendant officers at work. 

Second, the suspicion comes again that there may be a 
satirical intent underlying the account. There is so striking 

[ 190 ] 



CHANG HENG ON THE EASTERN CAPITAL 

a contrast between the initial emphasis on the necessary moral 
rectitude and the immensely detailed description of the pro- 
cessional trappings. And we know what little moral rectitude 
Emperor An showed when he began to get free of the 
Empress-Dowager Teng. 

For the meaning of the term t^en (heaven) in this con- 
text our objective is to ascertain what Chang Heng and his 
contemporaries thought about it. For the wider setting of 
the problem the student must go, for example, to the Ku 
SMh Pien ("Symposium on Ancient History"), VII, in par- 
ticular Vol. B (1941), where the various congeries of high 
ancestors, eponymous heroes, gradually-personalized totems, 
and sky gods and earth gods of the Sinitic and non-Sinitic tribes 
are to be seen in process of rationalization by successive steps of 
mythological interpretation. All this is germane to the main 
enquiry but cannot be dealt with here beyond this one ob- 
servation: the Han scholars struggled hard to make sense 
of it all. One result was the classification of the gods as 
either shen (gods in the sky), or ch^i (gods of the earth). 
This is plain from numerous examples in mid-Han memorials 
to the throne, as also in the "Record of Rites." The Five 
Shang Ti were major gods in the one category, Hou T^u 
("Ruler Soil") a major god in the other category. Further, 
the term ^^chiao ssu occurs again and again in relation to 
sacrifices distinct from those made in the ancestral fanes. 

The intriguing thing about Chang Heng's use of the term 
I'ien here in this section is that in addition to the general 
appellation ssu tHen for a sky-directed sacrifice he used three 
other appellations for the object of sacrifice: Shang Hsiian 
("Mystery-on-High"), T^ai I ("Supreme One") and Ling 
Chu ("Spirit Lord"). Now the second of these is not only 
highly theistic in meaning but also points to the idea of a 
High God above all the other gods, and on the way to being 
associated with THen. That certainly requires scrutiny. We 
go back to Yang Hsiung and the "border sacrifice" he saw 
at Kan Chilian about 10 b.c. In his commemorative ju he 

[ 191 ] 



CRITIQUE 

also gave alternative names for the major deity worshipped 
there: Shang-Hsiian and T'ai I, but also Huang-T4en ("Au- 
gust Heaven"). That would seem to confirm the impression 
that the term tHen was on the way to being personalized. 

Third, with regard to the changing of the T^ai I and Hou 
T^u sacrifices from out-regions to Ch*ang-an, the "History 
of the Former Han Dynasty" is most strikingly lacking in 
informative detail. For instance, Emperor Ch^eng's procla- 
mation in 31 B.C. reads: "We have removed the T'ai I and 
Hou T'u altars to the southern and northern suburbs." So 
also in 13 b.c, when he reinstated those sacrifices at their 
original sites, there is no mention of change of ritual. Yet 
changes of that sort usually brought with them implementa- 
tion of new ideas on the meaning of the institution concerned. 
Presumably there were no striking changes. We can easily 
believe this, for Emperor Ch'eng, and after him Emperor 
Ai, were both nervous over what the effect on the two gods 
would be. This is reflected in the "Acts of the Emperors" 
round those years. In such a case the man ultimately respon- 
sible would feel chary about making provocative alterations. 
The Wang Mang record is a little more instructive. In a.d. 5 
when Wang Mang was in the saddle as "Regent," he re- 
turned the two sacrifices to the Capital and boldly claimed 
that he was "obeying the mind of Heaven." With this move 
and instructions to Liu Hsin to build a ming t'ang at Ch^ang- 
an for the Five Ti, he ushered in his new dynasty. That 
ming fang was erected and used (see e.g., Wang Hsien-ch'ien's 
Han Shu Pu-chu^ ch. 99, p. 6a). The new Emperor, being 
a member of the Wang clan and not the Liu, had a difficulty 
over the Han ancestral shrines, and had to seek other means 
for placating their fanes. 

The question is: Were changes made in the nature of the 
T^ai I sacrifice? With regard to the last Former Han emperors, 
the answer is presumptively Noj with regard to Wang Mang, 
certainly Yes. In what, then, lay the distinction between the 
two? First, we go to Yang Hsiung's Kan Ch^Uan Fu^ com- 

[ 192 ] 



CHANG HENG ON THE EASTERN CAPITAL 

memorating the sacrifice made to T^ai I at Kan Chilian by 
Emperor Ch'eng (d. 7 b.c.)- The description there notes 
carriages poised so as to roll away of themselves and carry 
the sacrificial virlus to the ends of the empire, the "fiery 
fragrance offered to T'ai I," "chosen shamans all crying at 
the gate of [heaven's] Ti, opening heaven's audience hall, 
inviting a whole host of gods: a medley of helpers bringing 
down on the altar masses of good fortune, heaped up like 
a mountain." That was the kind of sky-directed sacrifice which 
could and did take place in the last twenty years of Former 
Han. Unfortunately there is no such detailed information 
about the New Dynasty's^^ method of procedure. There is, 
however, this: "When he [Wang Mang] knew in himself 
that he was defeated, he led his ministers out to the south 
suburb." He there called on "August Heaven" to send down 
a thunderbolt and destroy him, if he had done wrong. As 
he did this, he wept to the point of exhaustion. 

Throughout four-fifths of the Former Han epoch there 
was neither a sacrifice to Heaven at the Capital, nor any 
proposal that that there should be one. Yet when Emperor 
Wu's T'ai I ("Supreme One") god was brought to Ch^ang-an, 
we find that the altar set up for him is denoted in Wang 
Mang's time as an altar to "August Heaven": also that in 
A.D. 22 the New Dynasty's defeated Emperor abased himself 
before "August Heaven." Before that regime came into being 
the main attested site for an altar to Heaven was on the peak 
of Mount T^ai (where the establishment of new dynasties 
was believed to have been reported in past ages). There is 
also this: Kuang Wu, an extremely religious, not to say 
superstitious man, on the day he consented to ascend the 
throne, put the propriety of his decision to the test by setting 
up an altar to Heaven-and-Earth and sacrificing on it. The 
following year he established it in his new capital, and thirty 
years later he sacrificed to Heaven in the south suburb and 
sacrificed to Earth in the north suburb. He also started build- 

^^ The name of the usurper Wang Mang's short-lived dynasty. 

[ 193 ] 



CRITIQUE 

ing a ming t^ang for the Five Shang Ti. "Filial Son Em- 
peror Ming" carried on with the procedure, and these rites 
became established practices of the Later Han court. 

In conclusion, there is nothing monotheistic in the whole 
story. Yet in those stressful mid-Han years something was 
being engendered, something was emerging, which was in 
the nature of a religious philosophy — one in which, as the 
contemporary memorials show, a new term came into use, 
t^ien hsirij the "sky-mind" or "mind of Heaven." The ani- 
mistic matrix, from whence came the sacrifice Chang Heng 
depicted, is plain to view. But so also is the rationally quest- 
ing mind, which in that age sought to extract sense and value 
out of the mumbo-jumbo of ordinances. As a fragment of 
the Po Hu T^ung puts it, "The sacrifice takes place once a 
year — so as not to be too intimate with Heaven — at a time 
when the yang-^md begins to reassert itself." And again, 
"The King sacrifices to Heaven according to the same prin- 
ciple as he serves his father" (Tjan, of.cit.j 6s'i).^^ 

8. Other Ritual Duties and Moral Allegorizing 
on Them 

Apart from picturing the imperial ritual year, Chang Heng 
had other weighty matters on his mind (see Sections 12-16). 
So, having described in detail the pomp of the sacrifice to 
Heaven, he treated other main functions in briefer fashion. 
Six of them are dealt with according to their time incidence. 
For the first three the narrative is clearly demarcated by 
changing rhyme schemes. In the treatment of the last three 
this is not the case, although four rhyme schemes mark phases 

^^ One of the reasons why great caution needs to be used in envisaging 
purifying innovations in the Later-Han sacrifice is that the Po Hu T'ung 
in no w^ay emphasizes it. It is possible that there was a section on it in 
the original version of the Po Hu T'ung and that this got lost} cf. Tjan, 
of.cit.f pp. 61 and 652. If the sacrifice to Heaven really did not figure in 
the decisions of the conference, it is possible that there were innovations 
proposed by the reforming party at court, and the conservative party (whose 
opinion prevailed at the conference) fought for nothing to be reported 
either way. 

[ 194 ] 



CHANG HENG ON THE EASTERN CAPITAL 

in the narrative. In the circumstances it seems best to divide 
the section into five subsections/^ The genius of the descrip- 
tive fu consisted in the vivid presentation of objects, and accu- 
racy of description took precedence over strict prosodic form. 
These fu are veritable gold mines to the exploring historian. 
But it must be borne in mind that the pictures are impression- 
istic and may omit features which the contemporary reader 
would have had clearly in mind. For instance in this section 
Chang Heng makes no mention whatever of the sacrifice to 
Earth at the half-year, a much more important ritual duty 
than the Emperor blessing the drinking of wine. 



The critique of Pan Ku's Loyang fu left the reader with 
certain doubts, in chief one with regard to the functional 
range of the Loyang Ming T^ang. The first question to be 
faced is whether the proceedings included, as the Po Hu 
T^ung implies, a formal audience of the high feudatories and 
officers at which governmental policy was promulgated. Chang 
Heng's account reveals no sign of such an audience, agreeing 
in this respect with Pan Ku's ode on the Ming T^ang. They 
speak only of sacrifices, to the Five Ti and to Kuang Wu. 
Now, if there had been on the notable occasion of Emperor 
Ming's inauguration of his ming t^ang in a.d. 59 so notable 
a proceeding as a promulgation of policy, then Pan Ku's 
contemporary evidence is wanting in this important aspect 5 
and Chang Heng thirty years later is found backing his ac- 
count. Not only so: there is no trace of such a proceeding 
in the "History of the Later Han Dynasty." Also, the na- 
ture of the buildings would make them inconvenient, if not 

^^ The first three rhyme schemes are a very mteresting- example of ju 
prosody, for the first and second come to life again in the middle of the 
third. So far it is prose narrative of proceedings leading up to the actual 
shooting. The portrayal of that comes under the fourth rhyme, and the 
poem having done that in one four-word couplet, goes straight to allegori- 
cal moralizing and continues doing so to the end of the section, bringing 
in en route the Wine Festival and Entertainment of Veterans. The passage 
as a whole is, therefore, an illustration of what Lu Chi called "c/;'/;;^" 
"poetic mood"). 

[ 195 ] 



CRITIQUE 

prohibitively so, for such an occasion — although this would 
not apply to the presence of a limited number of dignitaries 
as witnesses of the sacrifices. The only reasonable conclusion 
is that Pan Ku and Chang Heng are right, and what the 
Po Hu T^ung reveals is the existence of an influential body 
of scholarly opinion which had other views of what should 
be done in place of what was done. 

Second, with regard to the sacrifices, Chang Heng's lan- 
guage here is even more illuminating than that of Pan Ku's 
ode. These deities of the Western Ch4n culture are presented 
in a rationalized form, being identified with the "Five Es- 
sences" {Wu C king) J i.e. the five constituent forces of the 
universe.^* Nonetheless Chang Heng calls them ling ("spir- 
its," "gods") and describes four of these ling as "energetic" 
{mo) and "loyally warm-hearted" {yiln huai). We seem 
to be back in the animist's world. Yet Chang Heng's mind, 
as we have seen, was by no means naively animistic. His idea 
is, therefore, in the nature of poetic fancy, revealing him as 
able to stand apart from those solemn rites, indeed think of 
them in a witty mood. Since ";;^o" and "Aw^i" are common 
in the "History" Scripture in reference to faithful ministers, 
it may well be that that was just the way in which Chang 
Heng meant to present them. In any case the picture he 
presents, with the Han virtus of fire specially honored, is 
one of religious sacrifices conducted entirely with political 
ends in view. Whether gods or forces of Nature, they must 
be mobilized by precisional rites in support of the reigning 
order. 

Third, in regard to Kuang Wu, deceased, made f'ei^ as- 
sociate god with the Five Ti, Pan Ku had stressed this, and 
it is stressed in the "Acts of the Emperors" under Emperor 
Ming's second year. Here, then, in this politico-religious 
amalgam, old ancestor worship and new filial piety are also 
to be found at work. King Wu and Duke Chou had long 

1* For the "Five Hsing>' and "Five T^," see Y. L. Fung, A History of 
Chinese Philosofhy^ tr. D. Bodde, I, 159-169. 

[ 196 ] 



CHANG HENG ON THE EASTERN CAPITAL 

ages back done this for their father, King Wen. In recent 
times Wang Mang as regent in a.d. i had had a suburban 
sacrifice performed in which Emperor Kao was made f^ei 
with Heaven and Emperor Wen f^ei with the Five Ti (see 
Han Shuj ch. 12). Thus to elevate Kuang Wu, the creator 
of Han Redivivus, was a demonstration to gods and men 
that the new regime had faith in its Heavenly commission 
and proposed to stand on its own merits. 

B. THE ANCESTRAL SACRIFICES 

If every culture has its own distinctive rhythms of thought 
and emotion, then here is a Chinese one. Chang Heng's mind 
swings straight from the energies of the seasons being con- 
trolled by sacrifices J to the ruling geniuses of spring, sum- 
mer, autumn, and winter j to a consideration of the fervent 
filial heart of the Emperor moved by the sight of these 
products so that he takes pious thought of his ancestral manes 
and the high duty of nourishing them. So he must needs 
make sure in person that all is duly prepared: the victims 
and their accoutrements, the succulent meats, the soup, the 
dances, and the bells and drums, in order that all may be 
done with "enlightened decorum." Now, the whole passage 
is conditioned by this initial picture of the "fervently filial 
heart" and, being so, may be summed up as a typical exam- 
ple of that mid-Han sentimental idealism over filial piety 
which is so impressive a feature of the "Record of Rites."^^ 
There are three keys in which emotion is seen at work here. 
There are the succulent meats presented to those forebears, 
whose flesh and blood the sacrificers inherit. One of them 
acts as impersonator and wears his grandfather's clothes, and 
in his company they now sit down and eat. There is the spirit 
of potency those forebears possess, so that being well wined 
they can be counted on to "send down good fortune in 
masses." And there is the emergent filial-piety strain (the 
Way of Filial Piety), bursting its ancient animistic bonds 

^^ See especially Sacred Books of the East^ xxviii, Book 21 ("AT/ /"), 
Oxford, 1885, pp. 210-235. 

[ 197 ] 



CRITIQUE 

and seeing the filial son serving the dead with no ulterior 
motive but only because they are the sacred dead to whom 
life-long gratitude is due. 

C. THE PLOUGHING OF THE FIELD OF HEAVEN 

Another procession out from the palace and the city is de- 
scribed. The Emperor took the plough in his own state car- 
riage, lying between the driver and the guard on the right. 
The special carriage with the phoenix bells and the "azure- 
colored dragon horses" are mentioned in the Yileh Ling 
("Monthly Ordinances"), and the same applies. Indeed the 
whole account is redolent of the Yueh Ling, though no men- 
tion is made of the Emperor's toasting the dukes at the end 
of the ceremony. It is an affecting picture that Chang Heng 
paints: the Emperor, on the one hand, with his own hands 
preparing the soil in which is to be sown the grain for the 
sacrifices, on the other hand, sanctifying and encouraging 
the toil of his hard-working subjects. Chang Heng appar- 
ently had no criticism to make of the proceedings j he was 
content to let the noble symbolism convey its simple lesson 
to the far from simple minds of his readers. 

As everyone knows, and as, indeed, the classical literature 
of China amply reveals, there was this sense of community 
between prince and peasant. There is, also, the other side 
of the picture, fully documented by the records, revealing 
the ease with which the peasant could be, and was, oppressed. 
That Chang Heng himself was fully aware of this side is 
shown in the prophecies of doom which later sections of the 
fu contain. As for the history of this rite and its connection 
with a tribal society in which a mana-imbued chieftain or 
medicine man turned the first clod, that story goes beyond 
the scope of this enquiry. This much may be said: that just 
as an insistent urge towards rationalization reveals itself in 
the Later-Han Ming T'ang and ancestor worship, so also 
something of the sacramental attitude to life emerges to view 
in this agricultural rite. The semi-divine Son of Heaven who, 

[ 198 ] 



CHANG HENG ON THE EASTERN CAPITAL 

as one current saying put it, "served Heaven as father and 
Earth as mother," can only adequately foster his high-priestly 
spiritual potency by the sweat of his body — by working the 
plough in the fellowship of humble labor. As Chang Heng 
expresses it, "in the supplying of grain in abundance for the 
high ancestral and suburb sacrifices, there is of necessity [p] 
a consummation of thought in personal [physical] exertion." 
Again the Chi I ("Meaning of Sacrifice") in the "Record of 
Rites" is instructive: "Only a sage can entertain the high 
gods J only a filial son can entertain his parents." 

D. THE ARCHERY RITE AND ITS PRELIMINARIES 

One or two touches convey the impression of an eye-wit- 
ness's observations. Since Chang Heng had studied in the 
Imperial Academy, he would in the ordinary course of events 
have been able to see the rite being performed, if only at a 
distance. The most notable feature in his enumeration of 
the preliminaries is his introduction of four legendary worthies 
and of Fu Sang, a fairy island — possibly Japan. 

In his ode on the Pi Yung, Pan Ku only referred to the 
autumn use of it for the rite of Entertaining the Veterans. 
Chang Heng refers to that here, but gives much more at- 
tention to the Archery Rite held in the Spring. Neither of 
our authors refers directly to the fact that the Imperial Acad- 
emy (Ta Hsueh) was on the Pi Yung site (in the south-east 
suburb of Loyang, seven // distant from the city). Pan Ku's 
omission is the more curious, since he was eulogizing the Pi 
Yung as a whole. Chang Heng, on the other hand, was deal- 
ing with the ritual year, and the Academy was outside that 
picture. As a matter of fact, he makes an indirect reference 
to it in his "advancing enlightened virtue" (see sub-section 
E), for skill in archery was one of the tests for graduating 
from the Academy. 

The thorny problem of education in the ancient royal capi- 
tals does not concern us here, except to this uncontentious 
extent, that archery unquestionably was part of the Chou 

[ 199 ] 



CRITIQUE 

system of training for the scions of noble families. Apart from 
that, the evidence is strong that either King Wen or Duke 
Chou conceived the political importance of having the feuda- 
tories meet together regularly in the capital, there to take 
part in the royal ancestral sacrifices and engage in archery 
contests. The rules of these contests were part of a code of 
chivalry by which was inculcated a sense of loyalty and honor 
and mutual forbearance. The evidence of the Analects is im- 
portant in connection with the existence of archery contests 
in Confucius' lifetime. That the tradition survived among the 
Scholars into Han times is not open to doubt, but there are 
no references to archery contests or a Pi Yung in early Han 
records. After Emperor Wu, the idea of a Pi Yung was in 
the air, not only as a training ground for young aristocrats 
but with ever-increasing emphasis as the chief training centre 
for aspirants to the public service. It was, then, with the 
establishment of the Imperial Academy and the encourage- 
ment of the students to practice archery, that the traditional 
code took on a new lease of life. It became impregnated with 
the new mystic consciousness of music, whilst the archer's 
stance and general behaviour became indicative of the quali- 
ties required for the imperial service. Both these ideas are 
strikingly emphasized in the "Meaning of Archery" {She I 
in the "Record of Rites"). 

E. THE MYSTIC ARROW: THE SIGNIFICANCE OF THE WINE FES- 
TIVAL AND ENTERTAINMENT OF THE VETERANS 

The introduction of a new and final rhyme marks the 
swing of the poem's mood from narrative to appraisal — 
appraisal of the imperial virtus in action, and so, by associa- 
tion of ideas, appraisal of the Wine Festival and Entertain- 
ment of Veterans. The Son of Heaven is the centre of the 
picture, and his arrow's striking of the target is represented 
as stimulating the tender shoots of the grain crop. That state- 
ment on the face of it indicates untrammeled belief in mana 
magic, a mana-imbued king and his shamans able to transmit 

[ 200 ] 



CHANG HENG ON THE EASTERN CAPITAL 

their mana into a dead implement and transform it into an 
agent of miraculous power. 

We have seen the transformation of the Five Shang Ti 
into emblems of the five constituent forces in Nature and 
the sublimation of the Emperor's ploughing rite. Both of 
these were ceremonies with their roots deep in the soil of 
primitive animistic belief. In the case of the mystic arrow, 
we must turn first to the section on archery in the "White 
Tiger Hall Discussions." There we find this: "Why does 
the Son of Heaven in person practise archery? To aid the 
yang fluid in stimulating the ten thousand things. In the 
spring the yang fluid is small and weak, and it is to be feared 
that the [ten thousand] things, meeting obstructions, will not 
be able to come out by their own strength. Now in archery 
[the arrow proceeds] from the inside to the outside 5 it pierces 
and enters the solid and hard [target], thus resembling the 
bringing forth of [nascent] things. Therefore, by means of 
archery they are stimulated [in growth]." This is ration- 
alizing, but rationalizing of that strained, pseudo-scientific 
kind which characterized the speculations of the Modern- 
Script school of thought which dominated the White Tiger 
Hall Conference. So far in our study of Chang Heng's mind 
we have seen no reason to believe he belonged to that ob- 
scurantist wing of scholar opinion, or subscribed to its fan- 
tastic logic of symbolism.^^ And here in this passage, having 
stated the theory of the imperial virius in the imperial arrow, 
Chang Heng says nothing about the piercing of the solid 
and hard, but goes straight to "reveals the heart of integrity 
in a far-reaching parable" (yuan yil). A man who had faith 
in the miraculous efficacy of piercing of a hard target would 
not have spoken of it as a parable. Further, in the succeeding 
couplet the pattern of thought is entirely on the moralistic, 
sublimating level. There too an arresting image meets the 
eye, "fierce appetites of ravening greed" (^H^ao tHeh chih fan 

^« See Y. L. Fung, A History of Chinese Philosofhy^ tr. D. Bodde, II, 
ch. 2 and pp. 91-92. 

[ 201 ] 



CRITIQUE 

yii), T'ao t^ieh were the central mana-induclng patterns on 
the Yin bronze vessels. One would not expect a believer in 
magic to translate so dangerous a symbol into the language 
of fierce moral disapproval. 

Any initial suspicion of Chang Heng's rationalism is now 
almost gone. But there is still the fact that he introduced 
four legendary figures into his narrative, depicting them ap- 
parently as actually present and co-operating in the proceed- 
ings: to put it bluntly, strengthening the virtus of the rite 
by their age-old numinous powers. Here we recognize, not 
only that Chang Heng's language is entirely precise, but in 
the case of Ta Ping and Feng Hou quite extraordinary. 
These two legendary figures feature in the Yellow Emperor 
legend. On the evidence of Hum Nan Tzu (ch. i ) they were 
fairy drivers who by the power of Tao "rode chariots of 
cloud . . . galloped wildly in the dim and distant beyond . . . 
crossing frost and snow and leaving no trace, shone on by 
the sun and showing no shadow." Yet here they are repre- 
sented as gentling the Emperor's mettlesome steeds along — 
presumably to avoid jarring his muscles. Again it is difficult 
to believe that Chang Heng is being pedantically factual. 
Rather, it is the imagery of romantic fancy — and witty 
imagery at that. As for Po I and Hou K^uei, they figure in 
the Yao-Shun cycle as respectively minister of rites and music- 
master. They also appear in Table Eight of the "History of 
the Former Han Dynasty" as "Men of Wisdom and Under- 
standing." 

With regard to the peasants' great feast of the year, Chang 
Heng's picture is useful to the historian for two reasons. One 
is that in making respect for the aged a feature of the harvest 
feast he gives us a date-mark for the infiltration of the filial- 
piety motif into the practice of peasant jollifications. The 
other reason is that Chang Heng introduces the Emperor 
as himself with his bare right arm cutting up the meat for 
his aged guests. The Son of Heaven thus demonstrated that 
he treated his guests as filial piety demanded a man should 
treat his own father and elder brothers. This feature does 

[ 202 ] 



CHANG HENG ON THE EASTERN CAPITAL 

not appear in the passages in the "Record of Rites." It does, 
however, appear in the "White Tiger Hall Discussions," 
where for the first time in extant Han literature there is the 
technical definitive phrase "three lao ["aged"] five keng 
["veterans of public service"]." It seems clear, therefore, that 
it was the Later Han emperors who began this nobly con- 
ceived ritual act of humble deference. Since in theory what 
was done at court was a model for the kingdoms and com- 
manderies we get a side-light on the proletarianization of 
the filial-piety doctrine. 

9. The Later Han Hunt 

Comparison of Chang Heng's account with Pan Ku's re- 
veals a general identity of purpose, namely emphasis on the 
Loyang hunts being a vast improvement on the Ch^ang-an 
ones. They may be suspected of partiality in favour of the 
existing regime, but even so there can be little doubt that 
the more barbaric spirit of the old sanguinary debauches was 
purged to a considerable extent. Both our witnesses are so 
almost naively exhilarated by the new spirit of wen (refine- 
ment) which they acclaim. Their evidence also agrees in 
making these Later-Han hunts more distinctively occasions 
for army review, and that with insistence on a high standard 
of efficiency in timing all sorts of deployment. Chang Heng 
had nothing to say on the use of cavalry, but then Pan Ku 
had nothing to say on Chang Heng's central point, the im- 
perial virtus reinforcing its power and appeal through the 
Son of Heaven's demonstration of humanity and mercy. In 

'^'^ The term ^^san lad*'' was current in the Ch'in First Emperor's and Em- 
peror Kao's time, but in a very different setting. Men of over fifty years 
were selected by the government to take responsibility for the law-abiding 
behaviour of their respective districts. The term "iu« keng''' seems to have 
been a later invention. According to Cheng Hsuan (cf. Tjan, of.cit.^ 479) 
it, as also san lao^ denoted one person, the '•'•san''' and "w«" referring to 
the two men's special qualifications. Ts'ai Yung, however, maintained that 
keng stood for sou (^^aged^^) and that there were actually three ho and 
five keng. Chang Heng's use of kuo ("state's") sou strengthens the student 
in believing (a) that keng in this connection indicates veterans of the pub- 
lic service, (b) that two kinds of old men were selected in Later Han times, 
three on grounds of great age, five on grounds of merit in the service of 
the state. 

[ 203 ] 



CRITIQUE 

this connection, as appeared with Chang Heng's moving pic- 
ture of the poor escaped pheasant and hare, he reveals a 
tenderness of heart which is very rare in the classic cultures 
of the world. 

With regard to "three drives," limited in number so that 
a way of escape was left open, Chang Heng seems to make 
an association of ideas between that and Sage-King Ch'eng 
Twang's net with three closed sides and the fourth left open. 
What the poem says is "He [the Son of Heaven] admires 
Ch^eng Twang's loosened net." One of the Later-Han em- 
perors may well have made an explicit reference to Twang's 
net in some order for the hunt, but the language of the poem 
may just as well represent nothing more than Chang Heng's 
poetic fancy. The point of the net story comes out in a pas- 
sage in the third century b.c. LU Shih Ch^un ChHu where it 
is stated that thirty southern countries hearing of the new 
king's concern for the fish inferred a like concern for human 
welfare and came and submitted.^^ 

The King Wen legend comes in the story of his life in 
the "Record of History." There it is stated that King Wen, 
being Lord of the West, attracted the notice of Lu Shang, 
the great Lord of the East Coast. The Lord of the West went 
hunting, the oracle being that he would capture neither 
dragon, tiger nor bear, but capture support for attaining the 
supremacy. In the hunt the two lords met at Wei Yang and 
in conversation Lu Shang was greatly pleased with the Lord 
of the West. It was a handy story for Chang Heng with 
which to point the moral. 

10. The Driving Away of the Demons 
of Plague 

The opening words make this part of the Emperor's 
priestly year, but as a matter of fact he is only in the back- 
ground, as far as the narrative is concerned — unless by any 

^^ For earlier evidence of humanitarian feeling for animals, see Mencim 
I A 17, which describes the sympathy of King Hsuan of Ch'i (331-313 
B.C.) for a sacrificial ox. 

[ 204 ] 



\ 



CHANG HENG ON THE EASTERN CAPITAL 

chance the sentences after "Thereafter" should be taken as 
having him as the unspoken subject. The objection to that 
is that Chang Heng so plainly is in lighter mood, and it 
is not likely that he would make His Majesty the subject of 
those ironical sentences. The impression of irony piles up, 
beginning with the couplet about the children's shooting and 
the demons being sure to fall dead. Then in the very next 
sentence they are racing away into the blue, and our poet 
piles Pelion upon Ossa with his beheading, braining, impris- 
oning, drowning, destroying, and exterminating. To crown 
all, after he has got the two beneficent Tu Shuo guardians 
on the watch, he says the lictors will see to any demons left 
over. In a word, when the holes and corners in the houses 
have been cleaned up, then nothing could be there "which 
was contrary to right and to rule." Moreover, the treatment 
is reminiscent of the Kan Ch^uan Fu in which Yang Hsiung 
had described the discomfiture of hostile demons: 

The mountain demons could not propel themselves here 3 
halfway on the long road they fell back. 

One after the other they fell headlong, the bridge of 
their flight broken, they became a cloud of midges and 
in a stroke were gone from the sky. 

In another part of the poem he wrote: 

The whole tribe of ch^i yu bearing fabulous swords and 
jade axes fled into hiding, racing away belly to earth: 
all, all of them, jammed together like weeds in an inter- 
locking mass, a rack of stormy clouds whirling away in 
wild confusion. 

With their ranks joined and overlapping, huddled to- 
gether they moved like fishes with eyes to the fore, like 
birds with necks outstretched. 

What marvelous descriptive imagery and superb irony! And 
it was for this kind of writing that Chang Heng's compeers 
had so cultivated a taste. 

[ 205 ] 



CRITIQUE 

With regard to the procession through the streets, again 
it looks as if Chang Heng had been an eye-witness. And 
this rite of expelling the demons of disease was part of vil- 
lage procedure, so that he may well have taken part in it 
as a child. This is the rite which Confucius was said to have 
seen, and which caused him to put on his court robes and 
stand to sympathetic attention (see Analects, X). The paral- 
lel is worth noting, for, surely, there is more than irony in 
those demons being bound to fall down dead. It is the child's 
process of reasoning: with these thousands of arrows shot 
into the air, how can the demons fail to get killed! 

That pathetic touch and the grizzly list of bogies' names 
brings home to us the fears which beset the lives of the 
Loyang populace. Historians know well what the cloacan 
filth of Paris and London was, even in post-mediaeval times, 
and the devastating epidemics which resulted. The point here 
is that one of the functions of government whether at a tribal 
or at a great-state level, is to take measures with the panic 
fears which seize a community on these occasions. So here 
in the poem there is put before us both the mobilization of 
the city government to deal with the emergency and the 
mimic warfare of the old animistic rite. The last was neces- 
sary, whatever the more highbrow scholars might think about 
its efficacy. On the other hand, the common sense of hygiene 
would appear to have been operative in Chang Heng's Lo- 
yang as well. Considering what the Loyang temperature in 
midwinter is now and was then, considering also the babies 
and old people in the families, the yearly official clean-up 
referred to in the last line would appear to have been of 
plain necessity. 

1 1 . The Tour of Insfection 

The Son of Heaven is back in the centre of the picture — 
how could it be otherwise with the subject what it is, the 
tours of inspection in state visits to the different parts of 
the country. Pan Ku had had nothing to say on this, and 

[ 206 ] 



CHANG HENG ON THE EASTERN CAPITAL 

now Chang Heng fills in the gap: rightly so, since these 
wide-flung visits were a vital corollary to the virlus theory. 
In terms of the Yin-Yang forces in Nature it was necessary 
that the sovereign, after gearing them into harmonious op- 
eration in the capital, should in person go and stimulate them 
under the different regional conditions. That a tour should, 
as Chang Heng's list of items shows, include action in the 
mystical sphere and on the practical level was of the very 
logic of the virlus theory. The Son of Heaven's influence 
naturally had to emanate from the capital, but it was neces- 
sary that he should also shed the light of his person in every 
quarter of the realm. That this entailed the labor and dis- 
comfort of long-distance travel was supporting evidence of 
the sage quality pertaining to the imperial government. 

It is clear from the treatment of the subject in the "White 
Tiger Hall Discussions" (Tjan, Po Hu T^ung^ p. xix) that 
these royal progresses stood high in public estimation in 
Later-Han times. The impressive thing in the two docu- 
ments is the highly bureaucratic nature of the Emperor's 
program on a visit. For instance, price lists of the goods in 
the markets were to be presented for inspection (Tjan, of.cil., 
496), whilst the poem mentions the planning of the crops, 
estimation of the harvests, and rewards for industrious sur- 
veyance. That there was a disciplinary side is emphasized 
in both documents, but the dominant motive seems to have 
been to introduce a debonair monarch among his loyal sub- 
jects, exhibiting his all-embracing human-heartedness. To 
judge from the general history of monarchies, eastern and 
western, ancient and modern, it has always paid sovereigns 
and their representatives to advertise and stage these popular 
occasions. Although in England the curing of the king's evil 
on a royal progress is a thing of the past, who dares say the 
launching of a battleship by a royal hand does not contain 
a lingering belief in a monarchic mana which can bring good 
fortune in its train? There was good statecraft in Mencius' 
warning to King Hsuan of Ch4, when he pictured the king's 

[ 207 ] 



CRITIQUE 

subjects saying, "Ir our king does not come journeying, how 
shall we get blessings? If our king does not make excursions, 
how can we get help ? " 

Another feature of the "White Tiger Hall Discussions" 
is the markedly feudal terminology used in connection with 
the tours. The intention in this is obvious, to underline the 
institution as one dating from sage-king practice. There seems 
no reason to doubt that it did exist in Western Chou times. 
For example, the retention in Later-Han times of the archaic 
term shou ("a hunt") points to a feudal society in which 
one of the king's functions was to visit his feudatories and 
encourage them in the destruction of beasts of prey. There 
is, however, ground for hesitation in accepting the tour of 
inspection described in the "White Tiger Hall Discussions" 
as applying equally to Former-Han times. In the "History 
of the Former Han Dynasty" the royal progresses do not 
figure in the early Han reigns as much as they do in the 
later reigns. Emperor Wu is the key figure here with his 
passion for traveling and his thirst for sacrificing in all parts 
of his dominions and so for proving that he had all the gods 
marshalled in his support. He had no compunction in taking 
unprecedented lines of action in his sacrifices at Mount T'ai 
and in the establishment of a centrally-appointed bureaucracy 
in every kingdom and commandery. The historian, there- 
fore, has to envisage the likelihood that during the succeed- 
ing reigns a new type of royal progress came into being, in 
which more exact methods of inspection were inaugurated. 
Unification and standardization, we know, became effective 
watchwords J and from this may be inferred a strengthening 
of the conviction that the sage-kings had foreshadowed these 
changes, had, indeed, conducted tours of inspection after the 
new pattern. 

Chang Heng's reference to the comfort of being back home 
seems a little pointed, particularly as Yang Hsiung in his 
Kan Ch^ilan Fu had written : 

[ 208 ] 



CHANG HENG ON THE EASTERN CAPITAL 

With the business finished and the merit colossal, we 
all . . . returned home . . . and took our ease in the 
Chiang Li Palace. 

The heavenly quarters were well taken care of, the 
bounds of earth laid open ... all the states in the world 
in peace and amity. 

So here, the complacency of the court having been pinpointed, 
a suspiciously elaborate description of the felicitous omens 
follows. As a matter of fact, the "Acts of the Emperors" 
shows that the omens during Emperor An's manhood years 
were by no means unfailingly auspicious. But Empress-Dow- 
ager Teng was surrounded by yes-men, courtiers who might 
be described, like the tsou-yil horses, as docile and "obedient 
to favors." 

The whole section is in fact patently hyperbolic, not least 
in its triumphant conclusion. There the peaceful virtus is rep- 
resented as a coverlet under which reposed the furthest of 
far-off lands. Ting Ling in Siberia, Southern Annam, Ta 
Ch'in (the Roman Empire), and Korea. We have to assume 
that Chang Heng knew something about the bloody cam- 
paign Tou Ku and his generals in a.d. 91 had waged against 
the Northern Huns, from which sprang Chinese contacts with 
the Ting Ling people. So also with Pan Ch'ao's plotting and 
violence in Western Asia, from which sprang contacts with 
Rome's frontier provinces.^^ It is when Chang Heng is in 
his particularly hyperbolic vein that the suspicion of irony 
cannot but assert itself. In this connection it should be noted 
that this section is the climax of the virtus argument on its 
noble side. The succeeding sections are patently critical. He 

^^ For details of the Later Han frontier wars and achievements, the 
memoirs on the barbarians in the "History of the Later Han Dynasty" 
(Hou Han Shu^ chs. 75-80) are a mine of information. Cf. E. Chavannes, 
T'oung Pao, vols. 5-8. For contacts with Rome, F. J. Teg-gart's Rome and 
China (University of California Press, 1939) is useful up to a point. For 
other works of reference see K. S. Latourette's The Chinese^ 3rd edn. re- 
vized, pp. 140-141 (New York, 1946). 

[ 209 ] 



CRITIQUE 

is ironical in a noble, poetic spirit directed at making people 
face their conventional sentiments in their naked reality. 

12. The Peofle^s Argument with the Throne 

The tenor of the poem changes at this point. It becomes 
more argumentative. With this change Chang Heng's drch 
matis fersonae^ the Teacher and the Nobleman, come more 
to life, and the two regimes, Ch'ang-an and Loyang, are 
made by contrast a burning subject of disputation. Teacher 
Anonymous first lays down the law, enunciating precedents 
which must be taken as principles of action. The removal of 
a capital must be after the pattern of the Shang dynasty's 
removal of its capital to Yin. This, as the P^an keng section 
of the "History" Scripture showed, was done "for the good 
of the people." So also in regard to luxury versus frugality 
in the court, the Ssu Kan ode gave the authoritative lead. 

The sacrifices at Mount T'ai and Liang Fu had not been 
subject to dispute in the ju but they had to come in because 
of their connection with the Yellow Emperor. In the Taoist 
and to some extent in orthodox mythology, he was the grand 
exemplar of government by mystic power of personality, and 
Chang Heng's prime contention is that the Han virtus is 
basically of that nature. That being so, it is not surprising 
that he uses the purest Taoist language here, particularly as 
he is not concerned with fiats, as Pan Ku was, but with the 
Son of Heaven himself. This personal note is struck all 
through the section, so much so that the Emperor is con- 
ceived as communing with himself. Yet it is surprising that 
the case should be presented in so Taoistic a form even 
though Confucius is introduced as being of the same mind. 
Chang Heng was writing for a highly orthodox court to read; 
and the salient fact is that he did so in this fashion. The Lao 
Tzu Book must in Chang Heng's eyes have had considerable 
authority as a Scripture on the Way and the virtus. 

Pan Ku's lyrical description of an imperial ban on luxury 
goods was found to have moderate confirmation in the rec- 

[ 210 ] 



CHANG HENG ON THE EASTERN CAPITAL 

ords. That applied to the reigns of Emperors Kuang Wu 
and Ming. Chang Heng was writing one to three decades 
later than Pan Kuj but though there was unquestionably a 
decline in morale, Empress-Dowager Teng was stiffly puritan 
in outlook and issued at least one edict against luxury goods 
(see Hou Han Shu^ X B). The phrasing in that edict re- 
sembles Chang Heng's here, so the question arises whether 
he had any particular court situation in mind as he wrote. 
The setting in the poem is curiously suggestive. 

First, there is that paean on the Han virtus in terms of 
all within the four seas being transported with joy over its 
scent "at this very time." Then, "only our sovereign lord 
can plant the auspicious ming chieh^'^ and then, "That being 
so, then how can the Tao not cherish, how can the [imperial] 
transforming power not soothe?" The sentiment might easily 
have come from the sycophantic lips of Emperor An's circle 
at the time when he was at last free of his domineering re- 
gent's will. Equally likely, a good old scholar — indeed Chang 
Heng himself — could make the same ejaculation, knowing 
that his sovereign's long minority had increased the sinister 
power of the eunuchs. But a rhetorical question can some- 
times be answered, and it was part of Chang Heng's thesis 
that the scent of the Han virtus had, at any rate in Ch'ang-an 
days, come to have a very mixed odor. However, all would 
be well if the Emperor proved himself equal to his oppor- 
tunity. 

The suggestion is made not only because of those particu- 
lar phrases. There is the poignantly expressed climax, the 
Emperor communing in his own soul, facing up to his Heaven- 
given responsibility, and seeing it as integrated with the winds 
and the life-giving rains. So also he faces the sage-kings. 
The Han virtus as embodied in the Loyang order went be- 
yond anything the Hsia, Shang-Yin, and Chou orders had 
achieved: it was equal to those of the Five Emperors before 
them. "Who says we have started too late and cannot over- 
take Fu Hsi and Shen Nung?" 

[ 211 ] 



CRITIQUE 

13. Sundry Reflections on Emferors and 
Their Way of Life 

Our attention is arrested at once (a) by the striking of 
an apologetic note, (b) by the resuscitation of the Teacher 
and his disputant, the Nobleman, who for so much of the 
fu have not been thrust on our attention at all. Clearly some- 
thing of special interest is to come, and the author for his 
part disciplines his thoughts by re-envisaging his dramatis 
fersonae. As to the apologetic self-deprecation, there are two 
separate remarks, the one clearly meant to be taken seriously, 
the other in its Teacher versus Nobleman context construable 
as an ordinary piece of politeness between two gentlemen in 
dispute. With regard to the first, since anything approaching 
to criticism of imperial policy necessitated the use of highly 
disparaging remarks about one's self, the natural inference 
is that the "something of special interest" is criticism of the 
existent regime. Yet on the surface Section 12 was purely 
eulogistic on the present Han virtus, and here in this pas- 
sage, although there are very pointed allusions to disgraceful 
incidents at Ch^ang-an, again there is still no overt criticism 
of the court. 

One is at first prepared to accept that Chang Heng had 
no hidden meaning here. But the passage as a whole has 
several extraordinary features, notably the zig-zag course 
along which the argument proceeds. Why should Chang 
Heng, in contrasting and condemning certain Ch^ang-an vices, 
have been compelled to jab out this straightforward subject 
in the way he does? And what had the theme of monarchs 
in their state carriages to do with those same vices, unless 
it was that the court Chang Heng knew was extravagant in 
this respect? In any case he does constructively condemn the 
court by urging that those immaculately trained carriage 
horses should be discarded. Working on the basis of that 
indication, again and again a double scope of reference, past 
and present, fits neatly to the statement made. Indeed, Em- 

[ 212 ] 



CHANG HENG ON THE EASTERN CAPITAL 

peror An seems to be the person hinted at, as, for instance, 
in "How much less can the inheritor of the imperial patri- 
mony treat his Heaven-given station lightly." Emperor An 
was very much under the domination of his favorite eunuchs. 
Also, one of them. Sun Ch^eng, is on record as being a man 
of violent temper and on one occasion at an audience "loudly 
berating those left and right." Compare "So [the lord of] 
a myriad chariots . . . may yet be intimidated by a com- 
moner." Sun Ch'eng had been ennobled by then, but he 
was a commoner by birth. Again, "All day he [the Emperor] 
does not leave his protective entourage — if he makes incog- 
nito excursions, where does he go?" True, that cap fitted two 
Ch'ang-an emperors, and, true, there is no actual evidence 
of Emperor An making incognito excursions, but in the light 
of his known weakly, dissipated character, and the Empress- 
Dowager's sternly moralistic training, what is more likely 
than that such excursions were made? Again Hsun Tzu's 
theory of education for "the One Man" was that he should 
be surrounded by exquisite sights and sounds and know no 
common, vulgar thing, and so have his character fortified 
against unruly passions. This theory was reinforced in the 
"Record of Rites" and Empress-Dowager Teng and her doc- 
trinal guide. Pan Chao, had a passion for the strict letter 
of the ceremonial law. So with regard to monarchs in their 
carriage and their sheltered rhythmic existence, does not 
the picture come near enough to a young man trained in 
the strictest school? Also, compare that picture, one of an 
exquisite dummy, with the counter picture Chang Heng 
also created of the Emperor communing in himself about 
his responsibilities. Compare also his picture of the two great 
emperors, Kao and Kuang Wu. They knew fear of danger, 
but not the pusillanimous fear of a gilded recluse who had 
never learnt to face the hard facts of imperial government 
and took refuge in sanctioning all that his sycophantic, ava- 
ricious councillors advized. It remains but to add that when 
Emperor An assumed his prerogatives he was ruthlessly ex- 

[ 213 ] 



CRITIQUE 

ploited by that eunuch gang, and when he died, the official 
eulogy recorded, "The virtus [of Emperor An] did not rise: 
it was [like] unripened grain in relation to our royal stand- 
ards." 

In considering this interpretative suggestion the reader 
should take into account the contexts to the three classical 
tags Chang Heng introduces into his argument, two from 
Lao Tzu and one from the Analects. Lao Tzu 26 reads as 
follows: "The sage man goes about all day and does not 
leave his tzu chung. Although he has palaces and harem 
buildings, his mind is aloof from them. How then about the 
lord of a myriad chariots who in his person treats the empire 
irresponsibly? If he is irresponsible, then he loses the root: 
if he is [meaninglessly] active, then he loses the sovereignty." 
Lao Tzu 46 reads as follows: "When there is Tao in the 
empire, then the fleet horses are used to dung [the fields]. 
I When there is no Tao in the empire, the war horses are 

^ reared in the suburbs. Of calamities there is none greater 

J than insatiability, of sins none greater than desire for gain. 

\ ..." In Analects y xiii 15 Confucius is asked whether one 

V yen (remark) can destroy a principality and replies, "Such 

•: an effect as that could hardly come from speech, but there 

r is the saying that *I have no delight in being a sovereign 

but only that my words are such that no one opposes me.' " 

14. Pla^n Facts of Social Economy 

The last section was highly poetical — vivid in its imagery, 
tense with restrained emotion, rugged and yet rhythmic. This 
section is the reverse : plain statements of plain economic facts, 
a cool enunciation in formal terms of how peace and pros- 
perity, the rational ends of government, come into being. 
In English the section may not seem poetry at all. In the 
Chinese these terse four-word couplets have the balance of 
poetic diction. There is the ring of their changing rhymes, 
and twice there is the alternation of rhythm through the 
use of six-word lines. Thus, although the passage is entirely 

[ 214 ] 



CHANG HENG ON THE EASTERN CAPITAL 

unrhetorical, it is eloquent, and above all limpidly clear, in 
all respects a poetic contrast to the passage it succeeds. 

This is rather a new side to Chang Heng. Although in his 
^^Fu on the Western Capital" he objected to the spawn in 
the streams being frivolously exterminated, he has not so 
far shown any particular interest in economic affairs. In point 
of fact we come here on a characteristic of the classical scholar 
mind, a marked streak of economic realism which can be 
traced back as far as Confucius (cf. e.g.. Analects^ i 5)> and 
which dictated much of Mencius' teaching. It is yet more 
clearly revealed in the Legalist philosophers, and even ap- 
pears in the Taoist mystical paradoxes. Leaving on one side 
the wider historical aspects of this feature as matter more 
pertaining to the final chapter, what needs noting here is: 
(a) Chang Heng's competent analysis of economic factors 
under two heads, natural resources and man power j (b) his 
emphasis on wise planning, if these two sources of wealth 
are not to give outj (c) with a flourishing economy the peo- 
ple are prepared to work hard and pay for more prosperity j 
(d) prolongation of this state of affairs leads to inter-class 
amity and loyalty to the throne j (e) fair-dealing becomes 
an established habit, and the ordinary man, illiterate as he 
is, responds to the appeal of integrity and self-restraint. 

It sounds very like the more cogent party speeches in a 
modern American election, or some of the patriotic appeals 
made in post-war Britain. In the China of Emperor An's 
reign this sober facing of facts was made in relation to a 
definite sag in the national economy. The prosperity of Em- 
peror Ming's time was fifty years back. And the public had 
reason for suspecting that intrigue and graft in Loyang were 
perverting the established safeguards of government effi- 
ciency. 

15. The Nobleman Forced to Face 

Unfleasant Facts 

In these nine and the succeeding ten couplets the poem 
reaches its climax, in fact the climax of the whole compo- 

[ 215 ] 



CRITIQUE 

sition. Since the double-harness mind worked so predomi- 
nantly in complementary effects, the right line of approach 
is to take the two sections as complementary to each other, 
the first culminating in an appeal to the Nobleman on selfish 
grounds, the second culminating in a general appeal on higher 
grounds. 

The dramatic dialectical setting to the poem is again thrust 
on the reader's attention: Teacher Anonymous, alias the au- 
thor, versus Nobleman Arrogant. The latter is straightly ad- 
monished, and whatever doubt may attach to Chang Heng 
having the Emperor and his circle in mind in Section 13, 
there need be no hesitation here in taking the Nobleman as 
representing a prevalent type among the aristocracy of the 
time. There is no beating about the bush. A denunciation 
is made in the form of two equations: 

1. The top-dog's oppression of the under dog 
= the under dog's inveterate hatred. 

2. The exhaustion of necessary commodities 
= rebellion and nation-wide misery. 

In algebraical terms Xi = Yi, X2 = Yoj and X^ -\- X2 
= Yi -|- Ys = -Z, and Z is a very uncomfortable time in 
store for "you. Noble Sir" and your friends. The logic of 
this is driven home, not by citations from holy writ, but by 
a succession of homely saws (images). The first, about water 
and a boat, was a current saying in that age. It dated from 
Hsiin Ch^ing, the philosopher, who made its scope of refer- 
ence doubly clear: as the water bears up and can equally well 
capsize a boat, so the people sustain the throne and can equally 
well overthrow it. 

The biting force of the other four images comes out when 
Chang Heng's sense of time is examined: time past, time 
present, time future, an ineluctable process. From his point 
of view it was a mechanically operative nexus of cause and 
effect in relation to human affairs, although when an effect 
would become operative was incommensurable. Thus, a hun- 

[ 216 ] 



CHANG HENG ON THE EASTERN CAPITAL 

dred years earlier, disastrous as Wang Mang's wicked usur- 
pation had been, the mysterious forces at work had taken 
time to mature, and it was twice nine years before the bogus 
regime crashed. So too with the mysterious but regular 
changes in the seasons j it is not till you wake up one morning 
and find frost on your shoes that you know hard winter is 
at your door. On top of that, the force of human nature must 
be taken into account — human nature with its natural pro- 
clivity to blind laziness. 

The quotation "If [you] rise at dawn . . . how much 
more . . ." comes from the Tso Chuan (Duke Chao, third 
year). These words were inscribed on a famous sacrificial 
tripod, and "how much more" was followed by "if daily 
[you] refuse to reform, [you] cannot endure [in the land] 
for long." Chang Heng cut out that trite moralistic warning 
to his noble readers — they knew it well enough as coming 
in the Scriptures. Instead, he aimed at the aristocrat's sense 
of bodily comfort and class superiority, accustomed as he was 
to robes cut regardless of expense. The day of economic dis- 
tress will surely come when a gentleman will perforce cut 
his cloth to a penurious pattern. 

1 6. The Final Word from Teacher Anonymous 

If the reader is expecting the eloquence of an impassioned 
appeal, he will be disappointed. Chang Heng's mood was 
otherwise. That the idea of eloquence occurred to him is 
evidenced by his reference to Ssu-ma Hsiang-ju and Yang 
Hsiung and their fu poems on the Ch'ang-an hunting park, 
in Chang Heng's day the applauded patterns of torrential 
eloquence. He glanced at that, recalling perhaps those of 
his friends who could smack their lips over the literary vir- 
tuosities and gently blink the appeal to action. He forces 
their attention to that by quoting the two poets' words, Ssu- 
ma Hsiang-ju appealing for the park to be given to the peas- 
ants to cultivate, Yang Hsiung calling for game preservation j 
and he crowns this by declaring that even if those measures 

[ 217 ] 



CRITIQUE 

had been taken, they would have been of no avail. Ch'ang-an 
was doomed, by reason of its ingrained vices. It did not mat- 
ter who ruled there, a descendant of Emperor Wu or a wicked 
usurper. Neither did it matter that it had strong passes to 
protect it. The day came when at the Han Valley Pass the 
government troops were beating their drums, and at that 
very moment the people's army was sacking the seat of gov- 
ernment. 

In a very real way this presentation of hard facts here 
is linked to the argument contained in the simile of the water 
and the boat. The people sustain the throne, and the people 
can overturn it: they did so a hundred years ago, opening 
the door for our glorious Loyang regime. Here is material 
for the historian. In the eyes of a Chang Heng the decline 
and fall of all-conquering, resplendent Ch^ang-an was the 
major fact in the country's history. That he used this fact 
at the climax of his appeal to the throne and its advisers and 
the nobility and scholar world demonstrates that it was for 
him the strongest ground of appeal. In other words, in their 
eyes, as to some extent in his. Later Han was so vastly su- 
perior to Former Han that, although there might be blem- 
ishes, the ship of state was travelling safely on its keel of 
enlightened rule and order. All through the fu Chang Heng 
has eulogised that enlightenment, and here he pins his hopes 
on being able to awake a commensurate enlightenment of 
mind in the guardians of the virtus. 

It was self-complacency he had to break up, in a blase 
young Emperor, in successful careerists enjoying very com- 
fortable emoluments. "It is human for the mind to think 
that what it has learnt is right, for the limbs to be comfort- 
able in habitual postures" ("but actually — I, Teacher Anony- 
mous, challenge you to deny it — there is a pungent smell 
about, and if you were not so used to it, you could not but 
be revoltingly aware of it"). On top of that thrust, the more 
caustic because of its market-place vulgarity, comes the refer- 
ence to the Hsien Ch^ih, the most refined of images in scholar 

[ 218 ] 



CHANG HENG ON THE EASTERN CAPITAL 

minds J for of all the court musical dramas which had come 
down from antiquity^° this was in theory the most purifying 
in its effects 5 and without music even //, the code of outward 
form, could not attain its ends. One of the hall marks of a 
scholar and a gentleman was that he responded to the Hsien 
Ch^ihh passion-subduing influence, as in his deportment the 
measured chiming of his jade girdle-pendants signalized a 
nobleman's dignity and honor. What did Chang Heng say? 
He said in effect: "You, Noble Sirs, are well aware that the 
court are bored with the Hsien Ch^th^ and would rather 
listen to the croaking of frogs, the base, libidinous music of 
Cheng which Confucius hated, as he hated those who with 
covetous mouths overturn states and families" {Analects^ xvii 
1 8 ) : "the great Music Master K'uang of such noble repute, 
he had a different mind, had he not: is he the only one?" 
The critique assumes that that question of Chang Heng's 
was both a rhetorical one and a genuine one. It was rhetorical 
since orthodoxy laid it down (see Tjan, Po Hu T^ung, pp. 
387-391) that from the Emperor down fine ritual and fine 
music were essential to them if they were to govern success- 
fully. The question also was a real one, for Chang Heng's 
hope was that every reader would take it to heart for himself. 
To the modern reader, peering through the long centuries 
at that patriot figure, brush in hand, his final appeal in this 
form may seem stilted and artificial, little calculated to strike 
home. It may be, for after all the appeal failed and the vices 
of that generation became worse vices and Chang Heng's 
warning of doom was most miserably fulfilled. On the other 

^° I take it that the historian can accept the evidence of the extremely 
learned Liu Hsieh (6th century a.d.), in Wen Hsin Tiao Lung^ ch. 7. He 
refers there to the music of the Yellow Emperor and his immediate suc- 
cessors as being "unascertainable for discussion." About what date did that 
and other antique music become obsolete and forgotten? In other words, 
is Chang Heng's peculiar reference to the Hsien Ch^ih music evidence that 
it was already on the way to being obsolete? It seems safe to assume that 
very antique musical dances were carefully transmitted, only with variations 
introduced by a succession of talented court producers. Cf. the tradition 
that Yao expanded the Yellow Emperor's Hsien Ch^ih^ made originally to 
purify his court of drunken and licentious orgies. 

[ 219 ] 



CRITIQUE 

hand, he knew his age, as he also knew the limits of the pres- 
sure an unennobled scholar might exert, and this was the 
form he chose. As has been emphasized above, sinners in high 
places were regarded as more likely to reform if skilfully 
encouraged to prod themselves.^^ 

17. The Nobleman Refents 

The lack of an elegant peroration to the Teacher's discourse 
has been noted. The poem as a whole has, to be sure, been 
amply eloquent, even bombastic in its Section 12. The re- 
markable thing is that here, after deploying his dry, caustic 
attack on blind self-satisfaction, Chang Heng continues his 
ironic line in the first couplet, turning his wit against himself 
and his prolixity, and then lets loose his eloquence in a pic- 
ture of penitence. He gives a portrait of a proud man tortured 
by the realization that he has both behaved badly and made 
a fool of himself. (One is reminded of the Hebrew poet's 
Job when brought face to face with the All-Righteous.) It 
was in this fashion that this Later-Han scholar and patriot, 
avoiding a vulgar display of his own fine sentiments, com- 
pelled his courtly readers to face their own souls. He could 
do no more, the rest was in the hands of fate — of Heaven 
that had brought Han and Later Han into existence for the 
good of the people, of the Yin-Yang cosmic forces which in 
man and nature unceasingly made the new old and so opened 
the door to recreation, with the wicked man destroyed in his 
wickedness. 

"^^ In Chuang Tzu (ch. 4) Confucius is presented as arg^uing with Yen 
Hui when the latter wanted to go and convert the arrogant and self-willed 
Prince of Wei. The sage warns the earnest young man against the danger 
of "forcing rule-of-thumb speeches about moral ideas upon violent brag- 
ging men." The well-meaning people who do that, said the Master, are 
rightly called calamity men, for they bring calamity on others as well as 
on themselves. Cf. E. R. Hughes, Chinese Philosofhy in Classical Times, 
Everyman Series, 1954 edn., pp. 186-189. 



[ 220 ] 



Chapter XI 



GENERAL APPRAISALS 

<><X><>0<><><X><><^<><^^ 

I. The Value of Pan Ku and Chang Heng 
as Witnesses 

I HAVE BEEN TRYING to achievc a vlgnettc of Han history 
through the study of two minds, related to each other by 
the fact that the two men lived for most of their lives in the 
same city, that one was born fourteen years before the other 
died, and that they both spent a great deal of energy in 
writing lengthy poems on the same subject. Since these two 
writers are my prime source of information, my first task 
is to appraise their value as witnesses. That is a question of 
relative intelligence and relative honesty. To my readers and 
myself who are in this case a panel of judges in a court of in- 
quiry, a pair of muddled-headed, unobservant witnesses would 
be useless j their respective stories would neither corroborate 
nor contradict each other. A couple of self-interested liars 
might be of some use, if we should be clever enough to get 
at the facts behind the misleading impressions they were out 
to convey. If, on the other hand, they were men of real in- 
tellectual acumen, with sharp eyes and attentive ears for 
what was going on around them, and if, better still, they had 
a habit of discriminating about what they learnt, whether 
it was fact or fiction, they would be invaluable j and the mere 
fact that at certain points their testimony disagreed might 
throw yet more light on the matters they describe. It is 
necessary, of course, to look out for signs of tendenz in their 
personal idiosyncrasies, their opinions as members of a class, 
and their unconscious assumptions as men of a particular age 
and particular cultural environment. 

In the foregoing chapters of critique, attention has been 
paid from time to time to these matters, and there is, in one 

[ 221 ] 



GENERAL APPRAISALS 

way and another, a considerable body o£ data. A summariza- 
tion works out as follows: 

(i) Pan Ku's record in regard to his Han Shu ("History 
of the Former Han Dynasty") has been dealt with in Chapter 
11. Certain signs of critical intelligence in that work are 
noted in the critiques. On the whole he is seen more at his 
judicial best in his historical work. In his two fu on the capi- 
tals he is more the prejudiced witness, not so much untrust- 
worthy about the bygone regime, his life-long field of study, 
but uncritically enthusiastic about the regime under which 
he lived, and to be suspected of not telling all he knew about 
the concomitants to imperial policy on the far western border. 

(2) Chang Heng, a mathematician and a deviser of pre- 
cision instruments, when he came to matters within the field 
of history showed himself about as penetrating as Pan Ku 
in his judgements on the past, and more trustworthy in re- 
gard to the political and social complexion of his day between 
twenty and forty years after Pan Ku's, when signs of a moral 
decline had begun to appear. 

(3) Both men were conscious of the unreliability of hear- 
say evidence, but both were not always consistent on this point 
when it came to universally accepted stories about the sage- 
kings of antiquity. Whether they accepted them literally or 
symbolically is not clear. If symbolically, then we need to 
remind ourselves that the history of thought is filled with 
instances of symbols achieving an aura of factuality. 

(4) Both men had a naturalistic philosophy, which to a 
considerable degree was superimposed on their inherited ani- 
mistic beliefs, transforming them somewhat but not dislodg- 
ing them entirely. 

(5) Both men's outlook was more cosmopolitan than na- 
tionally restricted, although both were persuaded that their 
own civilisation was fitted by its enlightened moral, social, 
and political principles, its cultured manners, its learning and 
literary genius, its command of resources and extension of 

[ 222 ] 



GENERAL APPRAISALS 

communications, to rule not only a greatly enlarged dominion 
but also all the lesser breeds without the law. 

(6) The four poems show a high level of literary com- 
petency. The authors were able to express subtle distinctions 
of meaning, to build up series of cogent propositions, to show 
genius in their use of apt and pithy illustrations, and to com- 
municate nuances of mood by the skilful placing of one re- 
vealing word. The four overall pictures they produced are 
composed of lucid representations of very complicated objects 
of sense-observation, and of their discriminating reflections 
thereon. In addition, whereas Pan Ku followed one order 
of consecutive portrayal and Chang Heng followed another, 
in all four cases the results reveal mature powers of coordi- 
nating effects and achieving artistic balance. And these quali- 
ties of the mind were shared, to some extent at any rate, 
by the circle of readers for whom our authors wrote. The 
outburst of literary production which Pan Ku's Preface so 
vividly depicts was accompanied by a new and excited con- 
sciousness of literary talent as one of the highest products 
of advancing civilization. Both Pan Ku and Chang Heng 
themselves, products of the new birth, were well aware of 
the critical minds that would study the fruits of their labors. 

(7) They both had a sense of humour, more of the urbane 
than the bucolic type, although Pan Ku could get near to 
slapstick. Chang Heng's command of irony was nothing short 
of masterly, so informed with wit, so straight to the mark, 
for the most part so free from the bludgeoning malice of 
wounding satire. 

(8) They both had powers of sensitive discernment in re- 
gard to emotional moods and physical disturbances, as, for 
example, the dizzying effects of flights of stairs (Pan Ku), 
and the intolerable ennui caused by a surfeit of routine en- 
tertainment (Chang Heng). They were familiar with the 
symptoms of mental confusion arising, for example, from 
nostalgia for a supposedly pleasanter past, or from conven- 
tional ideas making the unmistakeable evidence of one's senses 

[ 223 ] 



GENERAL APPRAISALS 

unbelievable. Chang Heng, at any rate, could appreciate why 
the peasant's kind of life made him satisfied with simple pleas- 
ures and then, when frivolously deprived of the necessities 
of life, made him blindly destructive in his hatred of the 
oppressors. 

(9) Although, therefore, their philosophic acumen came 
short of a modern logician's idea of precision thinking and 
Chang Heng's highest attainment in mathematics did not 
go beyond a faulty attempt at a more accurate reckoning of tt, 
yet along the main thoroughfares of human endeavour to 
achieve clarity of thought and precision of expression, to 
apply cause and effect on the cosmic scale and in the realm 
of ethics and social order, both Pan Ku and Chang Heng 
pass a general test of intellectual acumen. 

(10) As for their honesty of speech, this is more difficult 
of appraisal because of the sophisticated conventions of their 
age and the court they lived in. As the reader will have 
noticed, I have had my doubts about Pan Ku on the score 
of flattering his masters. But we have a double check on him 
in this. His comparatively low rank at court and his thirst 
for recognition were one thing: his trained eye for fact-find- 
ing was another. Of Chang Heng we get an impression of 
deeper sincerity. Faced as he was — in a way Pan Ku was not — 
with an ominous decline in court morale, he had a greater 
strain put on his sense of honor. His value as a witness is 
that much the higher because he nobly took the risk of ex- 
horting his imperial master and reproving the powerful no- 
bility. Yet he also used the terms of conventional adulation. 
We are, therefore, more able to appreciate the iron compul- 
sion of the code of manners and to avoid depreciating Pan 
Ku as a mere sycophant. Thus Pan Ku may be counted as 
medium honest: Chang Heng as superlatively so. 

The evidence from our two witnesses is, therefore, trust- 
worthy to a high degree, and even where it proves untrust- 
worthy, generally succeeds in being clear and precise. In a 

[ 224 ] 



GENERAL APPRAISALS 

word, these two voices from the past communicate to us. 
For example, as we have scrutinized Pan Ku's and Chang 
Heng's reactions to their environment, we have been able 
to appreciate those reactions. They make sense, in spite of 
the fact that the ideological values of Han China are in cer- 
tain respects alien to our dominant values. We are encour- 
aged to compare: to compare their "that" with the other 
"thats" of the world's cultural history and the "this" of 
our present experience. 

As for the relevancy of Han China to modern man with 
his need of usable knowledge, that is another question, one 
which can only be answered by surveying the data which 
the four critiques have assembled. Since politics and religion 
are two major fields of interest, they take pride of place. 

2. The Body Politic Set-uf: An Autocracy? 
A Theocracy? 

There is no attempt in the four poems to give a complete 
picture of the governmental set-up either in relation to Ch'ang- 
an or Loyang. Pan Ku comes nearest to it in his description 
of the old capital, but even that account is sketchy in relation 
to the chief administrative offices. From Chang Heng all 
we learn is that there was a practice of officers by rotation 
taking duty at night. About Loyang even that amount of 
detail is lacking. Pan Ku's description of the palace area is 
purely general, and Chang Heng's named list of buildings 
there leaves the historian groping in the dark as to which 
of them were government offices. In point of fact, both nar- 
rators were concerned with something else, not with the set-up 
but the source from which it derived its powers: the some- 
thing-else which might gear the machine to effectual working, 
or on the contrary might not. That was the Han throne, in 
their eyes a unique institution, unparalleled in the history 
of the race. In the first phase of its existence it had ended 
in calamitous failure. Only in its second phase was it a real 
success. 

[ 225 ] 



GENERAL APPRAISALS 

The basis of comparison was the Loyang monarchy. To 
both our witnesses it was a beautifully articulated institution 
which in dealing with Heaven and the gods, the Yin-Yang 
forces, the manes, the Chu Hsia (Sinicized peoples) and the 
whole world, gave for the first time free vent to the true 
virtus of government. Under that conviction the two poets 
wrote their respective Ch'ang-an poems. In those poems, after 
depicting the military impregnability and prolific economy of 
the city and area, they devoted their powers to pillorying 
Emperor Wu and his later successors. These are portrayed 
in general as irresponsible autocrats, free to indulge their 
megalomaniac and libidinous impulses. A special Han virtus^ 
brought into existence by the plebeian Emperor Kao Tsu, 
was evidenced by the continuity of authority and extension 
of sovereign power. But it was cribbed, cabined, and confined, 
unable to maintain efficiency and prosperity. How could the 
virtus be fully efficacious with vast expenditures and state 
sacrifices of unbridled luxury and determined by the hocus- 
pocus of shamans, and with the disciplinary side of govern- 
ment characterized by the cruel holocausts of the hunts? As 
Pan Ku emphasized, much of the Ch4n virus was at work 
in the body politic. 

I have argued that Later Han was essentially a theocracy. 
Former Han an autocracy. If to some modern minds this 
seems a distinction without a difference, what the critique 
has urged is that the Loyang theocracy was accompanied by 
features which suggest what we moderns call a constitutional 
monarchy. It should be noted then that the Vatican is the 
outstanding example of a theocracy in the West, and it also 
has over the course of the centuries developed a remarkably 
complete set of routine procedures. These are to all intents 
and purposes the constitutional law under which His Holiness 
the Pope exercises his theocratic prerogatives. In other words, 
his effectual authority depends on his discharge of his ritual 
and administrative duties and his recognition of the historic 
rights of the Catholic hierarchy. 

[ 226 ] 



GENERAL APPRAISALS 

It is the ritual of government which requires attention, 
for one of the surest ways of discovering the nature of society 
and its government is by examining its state rituals. In them 
is revealed in dramatized form the nexus of relationships 
which binds the society together. Now, the impressive thing 
about those Loyang rituals is that they represent so complete 
a combination of interlocking divine and human relationships : 
the Son of Heaven in compact with Heaven and Earth and 
the Yin- Yang forces, the Emperor on his throne in consulta- 
tion with his high vassals, the filial son serving the numinous 
dead, the monarch and the peasant in the fellowship of the 
sweated soil, the arch-governor and his scholars in the com- 
radeship of training for service, the exalted younger brother 
serving the nation's veterans. And Emperor Ming's reign 
was the first occasion on which all these symbolic acts were 
collectively performed in the capital for all men to note. 

The unifying principle which seems on all counts to com- 
bine these rituals into a coherent whole is that of the sover- 
eign's duty and responsibility. Emperor Kuang Wu had rights 
as a descendant of the High Founder and as the man of 
merit who had conquered all his competitors and brought 
peace. In him and his filial successor as occupants of the Han 
throne reposed all the time-honored prerogatives attached to 
it. Their word was law. On the other hand, they elected 
by the evidence of these rituals to exercise those prerogatives 
for the good of all their subjects. This was the Son of Heav- 
en's bounden duty in the sight of Heaven and manj and, 
to judge by the sage-kings and all the dynasties of antiquity, 
this was the only basis on which a dynasty could survive. 

That these rituals do not evidence the existence of a con- 
stitution is clear j but it is equally clear that such impressive 
year-by-year symbolic affirmations of principle point to there 
being something of the sort. So also does Emperor Kuang 
Wu's acknowledgement of economy as a cardinal rule for 
expenditure at court and among the aristocracy. 

The main evidence of constitutional procedure is to be 

[ 227 ] 



GENERAL APPRAISALS 

looked for in the field of government service and the method 
of recruitment for it. One of Kuang Wu's first acts was to 
follow his Han predecessors in establishing a corps of Eru- 
dites, fourteen in number. When peace came, the Pi Yung 
was built to house teachers and students, and the graduates 
were taken into government service. At the same time rec- 
ommendations of "the filial and upright" were obtained from 
the governing officials throughout the empire. Any man, 
however humble his birth, could count on his learning and 
ability being tested, and once his foot was on the ladder could 
rise to any height. For all this there was Former Han prece- 
dent. On the other hand, with regard to the hierarchy, the 
first picture of a comprehensively standardized bureaucracy 
comes to us from the Later Han records, namely Liu Chao's 
(Hou Han Shu^ chs. 34-38) version of Ssu-ma Piao's "Mon- 
ograph on the Hundred Officials." It is a most instructive 
document, covering as it does the whole gamut of imperially 
commissioned officialdom, at headquarters and throughout 
the empire. Every office, its title, duties, grade of emolument, 
is specified. Not only so: Liu Chao's foreword reveals Kuang 
Wu's reforming zeal at work, tightening up the service and 
establishing ch^ang hsien ("abiding rules of procedure"), 
whilst his commentary makes pregnant comparisons with For- 
mer Han.^ Add to this the enormous pride taken in Later 
Han in its enlightened // (the code of manners) by which 
one official recognized the rights of his superior and inferiors 
and their duties towards him. In these circumstances no other 
hypothesis fits the case except that of a bureaucratic state 
functioning with a nexus of reciprocative rights and duties, 
in fact a constitution. 

'^ The weighty nature of the evidence contained in these five chiian has 
not, as far as I have been able to discover, received the attention it deserves 
from modern scholars. Ssu-ma Piao (240-306) lived at a time when those 
court records which survived the Tung Cho incident in 190/1 were still 
in situ, and he had easy access to them. Liu Chao, living in that remarkable 
Liang Dynasty (502-556), seems to me to have had exceptional oppor- 
tunity for attaining a sound historical perspective, and to have used his 
opportunity to most illuminating effect. 

[ 228 ] 



I 



GENERAL APPRAISALS 

The question now is whether, in spite of Pan Ku's and 
Chang Heng's insistence that the two regimes were so vastly 
different, there is ground for believing that in Former Han 
times there had come to be much the same state of affairs. 
At first sight it seems that by Emperor Hsuan's reign (71-49 
B.C.) there was something very like a constitution in opera- 
tion. Thus Dubs has recently urged that there was a "virtual 
constitution" and adds an eloquent plea based on the throne's 
recognition of "Confucian" principles.^ Here it is, first of all, 
necessary to recognize that all or practically all of that round 
of rituals which Chang Heng portrayed had its prototypes 
in Former Han. That principle of the sovereign having duties 
and responsibilities to his subjects had since the days of Con- 
fucius and then of Mencius been part of the mental furnish- 
ings of the Lu State and Ch4 State Ju, Also Emperor Hsiian, 
and long before him Emperor Wen, had been admirable 
examples of jen and i (human-heartedness and justice). As 
for the bureaucracy, there is no evidence, that I have been 
able to discover, to show that Kuang Wu made radical changes 
in the system which had come down to him. 

There is this to be said for the Pan-Chang thesis. First, 
there is no question but that the Han emperors inherited the 
Ch^n ethos of government. As Ssu-ma Ch^en said, "the Ch4n 

" Dubs {History of the Former Han Dynasty^ H, S) refers to a "funda- 
mental unwritten constitution" as existent when Emperor Wu ascended the 
throne, and urges that he made changes in the direction of autocratic rule. 
That the first Han reigns saw the development of precedent in procedure 
is of course to be accepted on general grounds, if on no other; no gov- 
erning on a wide scale can continue without deliberately or undeliberately 
moving in that direction. But I have considerable doubts whether in so 
overtly a Legalist state as Former Han such a trend could have attained any 
settled proportions in sixty-odd years. The high officers of state knew that 
they were scrappable tools, and that masterful woman, Empress-Dowager 
Lii, held the reins for fourteen years. I grant, of course, that the disability 
of her sex led her to take the line of high consultation, but surely there 
was no guarantee of dynastic precedent about that. So I take it that Em- 
peror Wu ruled as he had a perfect right to do, not only in general but 
in particular in relation to his private secretaries and their controller. 
The hallmark of an autocracy is that anyone who has the ear of the 
monarch can stir him to override any and every precedent of procedure. 
Cf. Louis XIV's famous remark "L'etat c'est moi." 

[ 229 ] 



GENERAL APPRAISALS 

emperor lost his deer, and everyone went hunting it." The 
High Founder caught that deer, and it was his personal 
property, to do with as he willed. To his successors the em- 
pire was the sacred patrimony passed down to them for safe 
keeping. All through Former Han this was the unfailing 
theme, and although the idea of a mandate from Heaven of 
course came to be zealously preached, it was the only one 
amongst all the other religious struts to the throne founded 
on the Legalistic ground, right of conquest. Second, the Han 
Fei Tzu school of Legalist theory emphasized statecraft, so 
that the rulers' ministers were nothing but his tools, to exe- 
cute his commands, and, if the results were bad, to take the 
blame. Always the ruler must appear as the sole fountainhead 
of honors, rewards, public largesse, spurring his subjects to 
meritorious efforts, as by his published penalties and ruthless 
justice he destroyed the contumacious and drove the lazy and 
incompetent to increase the wealth of the state. This also is 
writ large on the pages of Former Han history. In A.F.P. 
Hulsewe's Remnants of Han Law (vol. I, Leiden, 1955) it 
can be studied in detail. 

Third, as to those religious struts to the throne, that array 
of time-honored high places where the Former Han emperors 
did periodic sacrifice, there was, as the critiques have stressed, 
the most intimate connection between them and a chieftain 
or ruler's achievement of mana authority. Here also the right 
of conquest, of possession of the soil on which the altars 
stood, is to be seen at work, not on the ground of any Legalist 
theory, but according to deeply entrenched tradition. The 
animistic mind was in command: the gods must be placated, 
could be placated, and indeed, as the Former Han emper- 
ors saw it, demonstrably were placated and acted in their 
favor. If calamities occurred, suitable measures were avail- 
able for restoring good fortune. Then, however, came Em- 
peror Wu and that vast expansion of empire. Nothing is 
clearer than that the situation demanded sacrifices of more 
extensive potency, and that Emperor Wu acted entirely ac- 

[ 230 ] 






GENERAL APPRAISALS 

cording to his autocratic will. His Ming T^ang sacrifices 
at Mount T'ai and the T^ai I sacrifice at Kan Chilian, etc., 
were a series of experiments, and from the point of view of 
his successors, on the whole successful experiments. As filial 
descendants entrusted with the imperial patrimony, they con- 
tinued to serve these altars, but with such alterations as they 
came to think were advisable for the buttressing of the throne. 

The alterations made were at first of little if any theological 
significance, since the dominant incentives to action were econ- 
omy, the convenience of the emperors in sacrificing, and a 
superstitious hope to increase the prestige of the Ch^ang-an 
administration. The struts to the autocratic regime needed 
strengthening, and these were the means adopted. On the 
other hand, to Wang Mang scheming for a throne it was 
of the utmost importance to placate influential opinion, in 
particular the scholars. Hence, with a mixture of motives 
including a belief in his own divine commission, his empha- 
sis on the south-suburb sacrifice as one to August Heaven. 
This move was attended by success both with influential 
scholar circles and in the good omens attendant on the sac- 
rifices. The result was a claim on his part to exercise un- 
trammeled theocratic authority. 

To sum up, it is clear that on its autocratic basis of the 
right of conquest the Former Han monarchy in time devel- 
oped some constitutional procedures in haphazard, uncoor- 
dinated fashion J also that all along there was a deep-seated 
theocratic element in the whole monarchic idea. To Kuang 
Wu and his supporters the problem was how to reinstate 
Han authority in a viable form, the autocratic kind of mon- 
archy being completely discredited by both gods and men. 
Here, I submit, the Pan-Chang evidence throws light where 
the extant historical records leave the historian bemused. In 
particular, Pan Ku's emphasis on the establishment of the 
basic human relations "for the first time" gives the essential 
clue: man bound by the very structure of his Nature-and-Man 
cosmos to live in a community of mutual service. The idea 

[ 231 ] 



GENERAL APPRAISALS 

was not new. It was there to be found in certain Scriptures, 
tentatively explored in certain Han pronouncements, but not 
for the first time solemnly affirmed as the foundation on which 
society rested. By this the throne pledged itself to consti- 
tutional practice. 

It may be argued that this spelt the triumph of "Confu- 
cianism.'" Rather, it is to be urged, that out of the Han maze 
of animistic beliefs and naturalistic logic, of autocracy tried 
and found wanting, there emerged a coherent view of the 
Great Society, man governed by a theocratic sovereign, and 
that sovereign pledged to constitutional practice. 

3. The Fact of Emfire and Its Ejfect 
on Men's Minds 

Were it not for Pan Ku's statement about the imperial 
hunts intimidating the barbarian world (Western Capital ju^ 
Section 9), the reader of the two Ch^ang-an poems would 
hardly realise that the earlier regime did enormously extend 
the sphere of China's power and prestige. Of course those ex- 
otic birds and beasts in the palace grounds and those blood- 
sweating horses from Ferghana were incidental indications of 
the expansion of empire. Also, there was no need for Pan Ku 
or Chang Heng to stress the facts with which their scholar 
readers were familiar, e.g. Chang Ch^en's diplomatic mission 
to Bactria and Li Ling's and Ch'en T'ang's resounding mili- 
tary exploits. Nonetheless the silence is noticeable, since it 
contrasts so emphatically with the lyrical exuberance of the 
Loyang poems in their depiction of the later world-enthralling 
virtus. Take also Chang Heng's sole reference to empire in 
his "F« on the Western Capital." His Nobleman in his sum- 
ming up refers not to the Former but to the Later Han 
emperors as having "covered the whole world and made one 
family." 

I submit that this contrast is of far-reaching significance, 
that after due deduction has been made for Pan Ku's and 

^ See H. H. Dubs, History of the Former Han Dynasty^ II, pp. 341-345. 

[ 232 ] 



GENERAL APPRAISALS 

Chang Heng's flattery of the court they served, there still 
remains a residual factj and that fact is that in their eyes 
whereas the Ch^ang-an court viewed the barbarian peoples as 
there for its greater glory and wealth, the Loyang court had 
a more far-sighted, more ethically statesmanlike view. It con- 
ceived of a comity of all peoples knit together under the 
Chinese aegis, their intertribal feuds composed, their bar- 
barous customs refined, the light of knowledge, technical as 
well as mystical, enriching their minds, above all participating 
in the Son of Heaven's embodiment of the Will of Heaven 
and felicitous ordering of the seasons. The sublime code of 
//, with all that it meant in pious sacrifices and the ordering 
of human relations, this was the possession of the Central 
State, and by the sharing of this inestimable treasure peace 
and plenty would ensue for all. 

The question is how this change of outlook came about. 
It has already been urged that the prestige and power of the 
Scholar bureaucrats in the last reigns of Former Han did not 
signify so definitive a doctrinal change of outlook as at first 
appears. At any rate the government's policy in relation to 
the barbarian peoples showed no sign of being activated by 
that code of reciprocal human relationship which Confucius 
had done so much to consolidate on an ethical foundation. 
It is necessary, therefore, to look for less doctrinaire factors 
in the situation. This search involves, first, a clarification of 
the component parts of the empire as it came to be in Former 
Han, second, an investigation of the emotional effect of the 
loss of empire concomitant with the fall of that regime. 

The spread of the Chou era's cultural influence is still a 
confused and somewhat indeterminate story. Yet it is clear 
that in its later phases some degree of Sinification of the tribes 
in the north and of the Yangtze Valley states, Ch^u and Yueh, 
had taken place. Then came the First Emperor's expedition 
to the far south, reaching to about latitude 25°, and those 
great areas came onto the map of men's minds. It was not 
until Emperor Wu's time that such far-flung conquests ex- 

[ 233 ] 



GENERAL APPRAISALS 

cited imperial ambitions, but from its early days the Ch^ang- 
an court was committed to establishing its authority over 
the surrounding semi-Sinicized areas. 

The pattern of control can be traced in outline starting 
with protectorates and resident Chinese advisers and leading 
on to full incorporation in the empire with the organization 
of commanderies with Chinese governors and officials, locally 
trained troops, statistical reports, and regularly imposed tax- 
ation. Also there were schools for the training of local talent 
as minor officials. It was from these areas, stretching from 
the Ch*eng-tu plain (Szechuan) to Kweichow and Kwangtung 
and reaching out in the north-east to Korea, that the Ch'ang- 
an government derived the main part of its unprecedented 
wealth. North, south, east, and west this inner empire, for 
all its variety of climatic conditions and natural products, was 
uniformly agrarian j and it was here that the Scholar officials 
were able to use their inherited wisdom to the greatest ef- 
fect. Defects there were of course, cases of peculation and 
oppression, but in general the public good was the officials' 
good J and where serious difficulties such as crop failures 
occurred, the central government was there to ameliorate 
the situation. 

In the north-west, the Hun area of influence and beyond, 
the situation was almost entirely different. There was very 
little scope for intensive agriculture, the indigenous inhab- 
itants living at the pastoral level, and the Huns in particular 
well able to contest Chinese interference with their tradi- 
tional way of life. Indeed, in its earlier days the Ch'ang-an 
government had bitter experience of their power to devastate 
the "Land within the Passes." Thus there was no peaceful 
penetration to the north-west. Either there was intrigue to 
set one barbarian faction against another, or great armies 
operating over terms of years, inducing temporary submis- 
sion. There could be no consolidation of empire, nor any 
reliable income from tributary potentates. There was money 
to be made if the oases on the long roads communicating 

[ 234 ] 



GENERAL APPRAISALS 

with Western Asia could be guarded, and this was the Ch'ang- 
an policy, never more than partially successful. Barbarian 
the regions were and barbarian they remained, enemies to 
be forced into homage to the Son of Heaven. A special class 
of officials came into existence, not Scholars of the Ju type 
but men from the border families and experts in barbarian 
politics, civil and military. 

What, in this milieu of empire building and the extension 
of hegemony, of the Juj the Scholars of the middle tradition 
who became so deeply engaged in governing and training 
men for the services? Again the historian finds a confusing 
situation, one in which a rigorous belief in Scriptural author- 
ity went hand in hand with violent controversy over the 
authenticity, meaning, and importance of any one text. Some 
light comes from Tung Chung-shu's (second century b.c.) 
synthesis of the "Spring and Autumn Annals" view of history 
with the Yin- Yang cosmology. That satisfied the minds of 
both Emperor Wu and the Ju as giving a rationale to "Great 
Han" in the world of peoples under the sky. A hundred years 
later came the Szechuanese thinker, Yang Hsiung, protesting 
against the fantastic speculations of the dominant school, yet 
himself indulging in abstruse calculations. In the last resort, 
however, his desire to enlighten took the form of a call to 
go back to Confucius, the real source of light on Nature and 
Man, for Yang Hsiung the sage of all sages to whom Great 
Han owed the Scriptures which it so meretriciously professed 
to obey. His Fa Yen ("Categorical Sayings") is the first solid 
indication coming from a known Han writer of devotion to 
the ethically minded Confucius of the Analects. His poetical 
writings reveal a profound disillusionment over the govern- 
ment of his day and their moral is unmistakeable : the Ch^ang- 
an virtus was not the vWtus of the sage-kings or the Scriptures, 
nor suited to world empire. 

It is the Pan-Chang evidence on which the historian must 
mainly rely for belief in a widespread sense of dismay at the 
malfeasance of the dynasty j but Wang Mang's efforts to com- 

[ ^ZS ] 



GENERAL APPRAISALS 

mend his new regime add confirmation. Most of the Scholars 
were anxious to be reassured and took office. Nonetheless, 
who could be sure that this was Heaven's decree? Only time 
could show, and what time did show was the crumbling of 
empire, ever increasing economic distress, and finally the sack 
of Ch^ang-an, the real idol of the Scholars' adoration. Then 
came the crisis : was the country of the sages and the Scriptures 
to sink back into the a-moral, suicidal disunity of the Warring 
States period, or was it not? It was a Kuang Wu who tipped 
the scale, and established what was essentially a new dynasty 
in a new capital.* Whether he approximated to Pan Ku's 
eulogistic description of him or not, there can be no doubt 
that he played for the Scholars' support and won it, nor that, 
when his throne finally became secure, he had the homage 
of the inner empire and of some regions of the outer. That 
he won this by the might of his military machine is unques- 
tionable, but on the other side stands the evidence of his 
purified sacrifices at his Altar of Heaven and above all his 
enunciation and enforcement of Confucius' great principle of 
economy. This he applied to himself and his harem, his court, 
and his nobles, and to all who drew emoluments from the 
public purse. This doubtless had high propaganda value, but 
it was the foundation on which ethical statesmanship could 
alone be built and applied to all peoples in the empire. 

Again what of the Scholars? Kuang Wu's Erudites, as far 
as we know, remained chained to their omens and prognosti- 
cations, and indeed their master was as anxious about this as 
they were. On the other hand, amongst the Erudites whom 
he appointed was Fan Sheng, a master in the Analects and 
the "Filial Piety Scripture," and in spite of the revulsion 
against Wang Mang's use of the Ancient Text Scriptures, 

* In Kuang Wu's struggle for power part of the trouble was the size of 
the peasant hordes which were in the field. Equally portentous were the 
rival claimants belonging to the imperial Liu and other clans. See Hans 
Bielenstein, The Restoration of the Han Dynasty^ Bulletin of the Museum 
of Far Eastern Antiquities, No. 26 (1953). The longest resistance came 
from semi-Sinicized Szechuan. 

[ 236 ] 



GENERAL APPRAISALS 

the Emperor insisted on having a chair for an expert in the 
Tso Chuan (see Hou Han Shu^ ch. 66^ Fan Sheng biog- 
raphy). His intention was finally blocked, but the biogra- 
phies of Cheng Hsing, Ch'en Yiian, Chia K^uei and Chang 
K^ai reveal that these devotees of the Tso Chuan and Ancient- 
Script texts did not hesitate to raise their voices at court. A 
new leaven was at work, something which Pan Ku acclaimed 
in the Preface to his two /«, something which inspired edu- 
cated men apart from the arid fields of textual controversy. 
The fact that stared men in the face was that "Great Han" 
had died, ignominiously, and "Great Han" was now alive 
again, gloriously — not next door to the capital of the Ch^n 
despots, but in Duke Chou's city of Loyang. What he ef- 
fected on the smaller scale among the feudatories could now 
be effected on the grand scale of empire. What was there to 
prevent the gospel of cultured manners and noblesse oblige^ 
of music taming unruly passions, of a law of magnanimous 
justice winning its way throughout the barbarian world and 
by this means bringing world dominion? 

The happy re-incarnation of "Great Han" in the new capi- 
tal and the return of the border peoples to allegiance syn- 
chronized with an awakened sense of a mission to the bar- 
barian world. The Yi, Man, Jung, Ti were not just enemies 
to be warred down and made a source of profit, but fellow 
humans needing the grace of the virtus to save them from 
the misery of intertribal strife and barbaric custom. As for 
the government's actual methods in Central Asia, it is doubt- 
ful whether this principle was put into practice to any effec- 
tive extent, but as far as the inner empire was concerned 
the work of integration went on. Year-to-year Sinicization was 
effectually accomplished in a dozen new commanderiesj and, 
when after a century and a half Later Han fell and its em- 
pire became three, one was centred in Szechuan and another 
near the delta of the Yangtze. Both were centres of vigorous 
and enlightened scholarship. The noble scholar family of Lu 
was a southern product: Lu Chi (d. dur. 196-220), the mathe- 

[ 237 ] 



GENERAL APPRAISALS 

matician, Lu Hsiin (183-245), the great administrator, Lu 
Chi (261-303), the author of the immortal fu, "The Art of 
Letters." And, going to the southern limits of the Han em- 
pire, there was that brilliant expounder of his Buddhist faith, 
Mou Tzu (flor. A.D. 200), a native of Chiao Chou (Annam), 
who said of himself, "I studied the Scriptures and Amplifi- 
cations and the Philosophers. I was a lover of books both 
great and small."^ 

4. Concerning Later Han Levels of 
Self-conscious Reflection 

The above two problems have been explored as a try-out 
in this task of appraisal j they seemed to stem directly from 
the main Pan-Chang thesis and to promise well-defined con- 
clusions. They have, as problems, proved difiicult to solve, 
and the validity of the conclusions is affected by incon- 
gruities in the evidence at our disposal. The outstanding 
difficulty lies in the apparently complete divorce between 
what our two witnesses affirmed with such noble eloquence 
and what absorbed the attention of the scholars and politi- 
cians of that time. Yet, when we compare the language forms 
of the Po Hu T^ung, for example, with those of the four fuy 
we find that the persons concerned were all products of much 
the same education and avowed devotion to very much the 
same sacred writings. The temptation is to label the Eru- 
dites of the Modern-Script school as obscurantist and Pan 
Ku and Chang Heng as either sycophantic or wildly idealis- 
tic. Resisting this temptation — as indeed a historian must on 
all counts — the task of appraisal drives him to closer con- 
sideration of this tangled and paradoxical situation. 

First of all with regard to Pan Ku and Chang Heng, the 
critique arrived at the conclusion that although their minds 
were stuffed with legends whose root and essence was purely 
animistic, they manifestly took those legends with a grain 

^ See the Foreword to Li Huo Lun in Hung Ming Chi, There is a trans- 
lation of Li Huo Lun by P. Pelliot: Meou-tseu ou les doutes leves, T'oung 
Pao, XIX (1920), 255-433- 

[ 238 ] 



GENERAL APPRAISALS 

of salt. This did not apply to the sage-kings, but then those 
stories had been rationalized and ethicized long before our 
two poets' day. They were hearsay, but hearsay guaranteed 
by special proofs. Nonetheless for all the dominant ration- 
ality of both our authors' minds, there were still pockets of 
incongruous animism lurking in them. This feature, the 
commonest of phenomena in the history of the world's ad- 
vanced cultures, represents the sub-conscious influence of an- 
cestral lore in the would-be rationalist j and it may in certain 
circumstances denote a restricted range of stimuli to self- 
consciousness, no more than simple powers of introspection. 
That kind of judgement, however, cannot apply to Pan Ku 
or Chang Heng, or to their more intelligent contemporaries. 

The four fu have revealed the clarity of imaginative vision 
and pungency of discriminate language of two first-class minds. 
What, then, of the second-class, third-class minds who pro- 
duced the mammoth Woof-Scripture literature? On its nu- 
merological side as also in its passion for lists of omens, this 
literature was the expression of a determination to link every 
kind of phenomenon, physical and psychological, human, ani- 
mal, vegetable, geographical, cosmological, into a coherent 
system of which the naturalistic Yin- Yang and Five Hsing 
(Elements) were the binding principles. The very extrava- 
gance of their numerological theories denotes a conviction 
that the world and everything in it can be reduced to mathe- 
matical calculation j and the correlation of portents was based 
on the belief that human conduct, being either ethical or 
contra-ethical, must have repercussions in the world of nature. 

There was, therefore, no fundamental dichotomy between 
the Pan-Chang type of mind and that of the less intelligent 
scholar-minds of that period. Both kinds could have existed 
side by side. Both exemplify the stimulating impact of "Great 
Han," unique Han, on their powers of thinking. Where the 
two types differed was that Pan Ku and Chang Heng, being 
able to exercise themselves in the freer atmosphere of poetic 
conception, looked for a gospel of world salvation. The main 

[ 239 ] 



GENERAL APPRAISALS 

bulk of the Ju had theological minds. They were dogma- 
tizers, fitting their new world to a jig-saw pattern of Scripture 
proof-texts J they wasted their ingenuity in a mass of techni- 
cal detail. Learned scholars can do that sort of thing, and 
see themselves as fighters for truth and the highest ethical 
standards. 

5. Concerning the Humanitarian Ethic 
and Its Origins 

For final appraisals it is not enough to allow a measure 
of Pan Ku's and Chang Heng's claim that the Loyang cul- 
ture was a humanitarian one. For one thing, rightly or 
wrongly, we moderns are unable to accept the theory of a 
divinely arranged succession of sage-kings operating from the 
beginnings of the race. For another, the humanitarian ethical 
idea is an ingredient in the great tradition of Chinese political 
philosophy, marking it as somewhat unique in the history of 
ancient cultures.^ It is the ABC of Sinological knowledge to 
recognize this as one of the supreme achievements of China's 
classical era. Since Chang Heng's "F^ on the Eastern Capi- 
tal" represents a high-water mark of this ideological achieve- 
ment, the problem here may well be envisaged in his lan- 
guage: "the dew of mercy is extended to bird and beast and 
insect, and the far-off places of the earth are saved" (Section 
9). The astonishing leap of his imagination from one affirma- 
tion to the other, and the astonishing fact that he should hope 
for some readers in the Loyang court to take the statement 
seriously! 

The historian's difficulty here, over origin and develop- 
ment, is that although the "ancient Scriptures" are explicit 

^ It might be claimed that this achievement was unique, but there were 
certain features in Ancient Rome of the Caesars which bear a close resem- 
blance; e.g. there was Hadrian's view of Roman citizenship as open to all 
dwellers in the empire, a view embodied in various codes. Also, imperial 
Rome envisaged itself as having a mission to serve the world. In the Jewish 
culture there is the humanitarianism of the Book of Deuteronomy, and in 
the later chapters of Isaiah the consciousness of a world mission to en- 
lighten the nations with the Torah. In both these cases the approach is on 
a different footing from the Chinese. 

[ 240 ] 



GENERAL APPRAISALS 

enough, there was that enormous gap of Former Han in 
which, according to our two witnesses, there was at most only 
fitful government recognition of mercy to men and the very 
reverse of mercy to animals. As a matter of fact, as all Later 
Han Ju recognised, the gap extended back into the period of 
the Warring States, an age in which the policies of all the 
states were flagrantly contrary to the example of the sage- 
kings. Further, not only have Chinese thinkers and writers 
subscribed to this view but also modern critical study has not 
produced any grounds for doubting that that age was a time 
of ruthless oppression and of great misery for the common 
people. On its positive side research has thrown new light 
on the causes of this state of affairs. In briefest summary, the 
mid-Chou era saw an increasing command of agricultural 
techniques and increasing accumulations of wealth in the 
hands of ambitious feudatories. The result was, first, a num- 
ber of swollen fiefs, achieved by conquest and absorption of 
neighboring fiefs, second, th-^. ruling of large bodies of serfs 
unattached by any traditional bonds of mutual services and 
mutual benefits. Right through the period of the "Spring and 
Autumn Annals" this process continued so that by the War- 
ring States period North China consisted of competitive inde- 
pendent states. The only feasible method of government was 
by rigid rules of taxation and published codes of legal penal- 
ties for crimes against the state. With the fierce competition 
that came between state and state, there was no room for a 
humanitarian ethic of government, whatever might have been 
the case in dead-and-gone feudal times. 

For the Chou feudal order in its heyday the evidence is 
almost entirely what is to be found in the "Odes." As a book 
this is riddled with interpretational conundrums, yet in its 
mass effects it is unmistakeably delineatory of the society 
which produced these three-hundred-odd poems. In them the 
harsher side of feudalism comes plainly to view: so also does 
that personal relationship between lord and peasant which 
is characteristic of a feudal order. It was a relation of patent 

[ 241 1 



GENERAL APPRAISALS 

inter-dependence, each party rendering services vital to the 
existence of all. 

This "Odes" collection of tribal songs, rural courtships, 
festal odes, personal laments from castle and cot, and temple 
chants was put together by Confucius in the last generation 
of the "Spring and Autumn" period. Confucius was a /«, a 
member of that intermediate class between lord and peasant. 
They were not Scholars in the full Han sense, some of them 
no more than shamans, but all of them repositories of tradi- 
tion. Somewhere in the traditions they guarded there was 
enshrined that consciousness of a personal bond between lord 
and peasant, each needing to be respected for the indispen- 
sable services he rendered : the lord to be honored and obeyed 
as the tribal chieftain had been, the peasant to be encouraged 
at his tasks and saved in his times of distress. 

At the end of the sixth century b.c. this was a live memory, 
the ancestral way. For the young Juy Confucius of Lu, it took 
shape in his mind as the divinely-ordained key to his coun- 
try's problems, as in sober fact the way King Wen and Duke 
Chou had elected to rule. The only remedy to the strife and 
disorder in the states was to "go back to Chou." The Master 
spent his life working to this endj and whether his vision 
of Duke Chou's benevolent statecraft was a figment of his 
imagination or not, is a question to which today there is no 
sure answer. In any case in his old age he realized that he 
had set himself an impossible task: the power-holders and 
power-seekers in Lu and elsewhere would have none of his 
dream. For his intimates, the younger men who entrusted 
their hopes of a career to his powers of training, it is impos- 
sible to say how many were under the same firm conviction 
as their teacher. Clearly they often had difficulty in grasping 
the full significance of his language. Yet unquestionably he 
had power in the minds of some of them, and through the 
dark centuries that followed their disciples treasured the mem- 
ory of his sayings. This "Way of Jen^'^ — that vague inclusive 
term of his which included all humane thought beyond the 

[ 242 ] 



GENERAL APPRAISALS 

conventional demands of family and clan relationships — be- 
came enshrined as the "Way of the Sage-Kings," and those 
small bands of Ju who cherished the Confucian tradition 
exalted him to be the last of the sage-kings, the "Uncrowned 
King." By early Han times that was part and parcel of the 
virtus residing in his name 3 and, since the concept of him 
thus became half -myth, the essential virtue of his teaching 
was the more subject to the farti-fris theorizing by the various 
schools of the Han Ju. 

Whether this reconstruction of the Confucius of history 
be sound in every detail is not to the point here. The ap- 
praising contention is that with regard to the high and holy 
duty of rulers to rule primarily for the good of the whole 
nation and with humane concern for the common man, Con- 
fucius of Lu was its main author and source of inspiring ap- 
peal. That, I submit, is not open to doubt. I also submit that 
Ch'ang-an saw itself as far removed from the days of simple 
feudalism. In a bureaucratic and legalized state there was lit- 
tle room for Confucius' dream. Loyang, also a bureaucratic 
and legalized state but born in a time of blood and tears, 
was for a time more responsive, and it was possible to en- 
visage the dream as at least one indispensable foundation of 
world empire. 

6. The Filial Piety Strain 

As a concomitant to the humanitarian ideal there stands 
in the four poems what has been described in the critiques 
as the eudaemonistic strain. That strain stands counter to what 
may be called the puritan type of ethic, the subscription to 
a moral standard solely because it is right. Mencius shows 
evidence of this, although not to the exclusion of the eudae- 
monistic type. The combination is, indeed, a feature of that 
intellectually lively and in so many ways progressive age, 
the last century b.c. It was part of its interior consciousness 
of the individual and his emotions, of the influence he can 
exert in his community, and of the austerity of personal dis- 

[ 243 ] 



GENERAL APPRAISALS 

cipline by which the man of breeding and a sense of honour 
must cultivate himself. Along with this went acute conscious- 
ness of the pleasures of the table together with a studied 
delight in music and the beauty of all forms of art, including 
that of courteous behaviour. Nowhere does this appear in 
more subtle form than in the "Way of Filial Piety" (Hsiao 
Tao), that distinctive achievement of classical Chinese culture. 

At first sight the Pan-Chang evidence would cause us to 
minimize the part that the Hsiao Tao played in the motiva- 
tion of the Former and Later Han folk. The same applies 
to ancestor worship. In Pan Ku's "F^ on the Western Capi- 
tal" there is no reference to filial piety and only one to ances- 
tor worship, and that is just by the way. Chang Heng's ^^Fu 
on the Western Capital" has no reference whatever. In Pan 
Ku's "F^ on the Eastern Capital" there is a bare reference 
to the seasonal sacrifices to the imperial ancestors. Chang 
Heng, in his calendar of the year's sacrifices, includes the 
one in the Loyang ancestral fanej and later he notes the 
routine autumn visit to Ch'ang-an to sacrifice in the High 
Ancestor's fane. He starts the former account in good filial- 
piety terms but winds up with what looks like an ironical 
touch on the blessings to be expected. 

To infer from this scarcity of reference that Pan Ku and 
Chang Heng were not conformist ancestor worshippers and 
staunch filial pietists would, to be sure, be wrong. So also 
would be any inference, from the exclusion of ancestor wor- 
ship from not only the Ch^ang-an and Loyang palace areas 
but also from inside the city walls, that these rites were not 
of central importance to the emperors. High-brow Scholars 
of the Pan-Chang calibre may have felt critical of the ani- 
mistic beliefs preserved in the rituals of ancestral fanes j they 
would not doubt the essential rightness of these memorial 
practices in the ancestral way. The problem that comes to 
mind here is that of relating Chinese ancestor worship, a 
type form of ancestor worship all over the world, and that 
special Chinese discovery of Hsiao Tao. It cannot be ignored 

[ 244 ] 



GENERAL APPRAISALS 

in the task of appraisal, for, of all the blessings of civilization 
which Later Han was persuaded it could impart to barbarian 
peoples, the cult of filial piety was unquestionably the chief/ 
Although the two phenomena are so intimately connected, 
it is confidently assumed that they are strikingly disparate, 
standing on very different levels of ethical aspiration. Ances- 
tor worship is impregnated with the fear of the dead and 
never succeeds in breaking away from the rudimentary logic 
that the tasty dishes the living enjoy will also be enjoyed 
by the ghosts of the dead, even though it be only by the 
smell. Thus the mana of the "old man," so potent in his 
lifetime, whether in favor or disfavor, must be even more 
potent after his death. The same applied a fortiori to the 
"grand old man" — the original progenitor of the tribe or 
clan. That this kind of mentality was in Han times vigorously 
alive in all ranks of society is completely evidenced, particu- 
larly with regard to the emperors and their entourage. The 
long and highly emotional struggle over the keeping or not 
of all the funerary parks demonstrates this for mid-Han 
times. On the other hand, the markedly mid-Han produc- 
tion, the "Record of Rites," contains works which are im- 
pregnated with a totally different attitude to the dead. In 
that group of three very freshly written dialectical essays 
(lun)y the Cki Fa^ Chi I and Chi T^ung,^ the great argument 
of appeal is that the sacrifices to the dead are made out of 
pure love and reverence, the filial son passionately longing 
for some sign that his adored parents are near. There is no 

'^ The C/ii I ("Principles of Sacrifice") in the "Record of Rites" has the 
passage: "Tseng Tzu said, *Exalt filial piety, and it includes heaven and 
earth J broaden it, and it encompasses the Four Seasj hand it to future ages, 
and morning and evening it will expand. . . .' " 

^ The C/ii Fa has no distinctive exposition of piety in relation to the 
sacrifices. It is entirely concerned with the proper rituals for the sacrifices 
to Sage Kings and other great figures of high antiquity. Nevertheless, it 
betrays the same kind of rationalizing mind as the other two lim, and in 
its final section asserts the principle of grateful remembrance of those wdio 
had long ago rendered outstanding service to man. The other two lun are 
vital for an understanding of Later Han filial piety. 

[ 245 ] 



GENERAL APPRAISALS 

better evidence of the cogitational acumen which the enquiry 
of this book has brought into clearer perspective. 

The history of the emergence of this Hsiao Tao is beyond 
the scope of this enquiry j it goes back to the early days of 
the Chou regime, and began to undergo some quite startling 
developments in Confucius' generation. Confucius' emphasis 
lay on serving parents when alive in homely acts of service. 
His disciple Tseng Ts^an went much farther, inculcating life- 
long remembrance of the dead. The tendency towards ex- 
travagance in funeral rites became more and more marked. 
Indeed, in the Warring States age only the wealthy could 
figure as filial sons. The downtrodden peasants had no an- 
cestral fanes, and often could not afford coffins at their par- 
ents' death. 

It is submitted that the final downfall of the Warring 
States entailed the ruin of their rulers' ancestral fanes and 
the proletarianization of their remaining descendants. Those 
of them who were literate maintained as best they could the 
aristocratic ritual tradition. With the rise of prosperity under 
Han the scene was set for a popularization of the Hsiao Tao, 
the Ju taking an active part in teaching the right forms for 
mourning and daily service. Then came the rise of the Ju 
to power, the establishment of schools in the provinces and 
contemporaneously the intensive study of the traditional rites. 
With the return of prosperity in Later Han and further dif- 
fusion of education, the moral teachings of the Ju penetrated 
more into village life, turning the hearts of parents and chil- 
dren to noble acts of ungrudging devotion. Pan Ku's eulogy 
of Emperor Kuang Wu speaks of the "Five Lun*'' (basic 
human relations) as his act of creation. Although sometimes 
in Han writings the relation of sovereign and subject comes 
first in the list, the dominant practice was to make the rela- 
tion of children to parents the chief duty of man. Filial piety 
became the categorical imperative. Pu Hsiao (stark contra- 
vention of the filial piety code) became the worst of sins — 

[ 246 ] 



GENERAL APPRAISALS 

an accusation of which made the most hardened sinner blench, 
from the Emperor on his throne to his meanest subject. 

7. The Sfread of Educational Facilities 

It may seem to some readers that the Pan-Chang evidence 
on education is of no great significance. That, as the critique 
on Section 6 of Pan Ku's "F^ on the Eastern Capital" has 
pointed out, is not the case. The statement there, the only 
one in the four fu which refers to education as such, is re- 
markable for a double emphasis: (i) on the empire-wide 
establishment of schools 5 (2) on the village school as a cen- 
tre for community sacrifices and feasts. In that connection 
the critique has made two possibly fanciful conjectures as 
to what lay behind Pan Ku's words. Apart from them, what 
he very explicitly did have in mind was to depict these gather- 
ings as a channel through which the imperial virtus flowed 
out to "all within the four seas." The village people, old 
and young, "tread the measure of te and chant of [the good- 
ness of] ;V;^," i.e. they learn the code of //' and are imbued 
with its concord-engendering power. It is the same process 
at work which has been described above in relation to the 
proletarianization of filial piety. Here this power is ascribed 
to the hsiang (high schools in the provincial centre) and hsii 
(village schools), to the ever wider extension of education 
throughout the incorporated regions of the empire. 

In view of the remarkable cultural influence exercised by 
China over the medley of tribes east of the Himalayas, it 
is necessary to see this proletarianization of education not as 
a figment of Pan Ku's imagination but as a process which 
did in some measure actually take place. ^ The question then 

^ For confirmation of this view see the Po Hu T'ung (Tjan, of.cit.^ pp. 
487-488) and take into account that according- to the Ju Lin Chuan fore- 
word {Hou Han S/m, Biog. 69A) there were about A.D. 146 "35,000 yu 
hsueh students" going from one teacher to another. The "Notes on Edu- 
cation" {Hsi'ieh Chi) of the "Record of Rites" gives tantalizing glimpses 
of provincial schools, but its main interest is in the higher education in 
the capital. Arguing from the general tone of the document and the incen- 
tive there was for commandery and kingdom high schools to ape the 

[ 247 ] 



GENERAL APPRAISALS 

is whether this advance must be dated as taking place during 
the Later and not the Former Han dynasty. A strict inter- 
pretation of the Pan-Chang thesis would lead to the conclu- 
sion that it did, since the Ch^ang-an court became too be- 
sotted with its own grandeur to be able to release the power 
of the true Han virtus. The evidence, however, does not war- 
rant so wholesale a conclusion. The most that Pan Ku and 
Chang Heng could have meant was that the Loyang em- 
perors had vastly increased the momentum of this develop- 
ment in education. It is urged that this actually was the case. 
The first major steps towards popularization of education 
must have synchronized with the implementing of Emperor 
Wu's policy of having better trained government servants 
!j and larger numbers of them. It became increasingly neces- 

;} sary, if a "filial and upright" candidate for office was recom- 

3j mended from the provinces, that he should prove on exam- 

•f ination to be reasonably well grounded in the Scriptures. That 

;', meant years of preparation under competent teachers, and — 

') so the critical historian inevitably surmises — the formal dis- 

\. tinction between higher grade and primary schools became 

]: more or less recognized. With the establishment of the Ta 

1 Hsueh (Imperial Academy) in Ch^ang-an and the employ- 

.. ment of the most famous Ju^ the Erudites, as teachers in it, 

i a new standard of scholarly competency was set up. Within 

\ fifty or sixty years there came to be on the one hand those 

\ Ch^ang-an street-corner Scholars "splitting hairs and dividing 

the muscles [of Scripture] asunder," and on the other hand 
the litterateurs, excited by the unexplored possibilities of or- 
dered expression. As early as Emperor Wu's reign there was 
Ssu-ma Hsiang-ju, educated in distant semi-Sinicized Sze- 
chuan, but able to electrify the court by his literary brilliance. 

methods used in the capital, it seems likely that passages in the Hsi'ieh Chi 
refer to those provincial establishments. It also seems likely that if one of 
those yu hsueh students was obliged to take charge of a provincial school, 
he might easily be guilty of the faults of "teachers today" as described in 
the Hsileh Chi: "humming over the tablets before them and propounding 



knotty problems. 



[ 248 



GENERAL APPRAISALS 

What, then, of the first hundred years of Han? To what 
extent were schools to be found scattered about the central 
states? The evidence is scanty in the extreme, and the exist- 
ence of scholastic centres is mainly by inference from the fact 
that there were famous Scholars and competent administra- 
tors. That they were really literate is obvious, although there 
is the teasing question what difference the new script had 
made to teaching method and how far Scholars were at home 
in it. Chia I of Emperor Wu's reign clearly had mastered 
it, but presumably when he went to Ch'ang-an he read the 
Li Sao in the old script. Altogether the situation is very 
much blurred and in no way clarified by the fact that the 
Ch'in First Emperor used a large number of clerks in run- 
ning his government offices. They must have learnt to read 
in the old script j but what standard of proficiency did they 
reach with its clumsy and variable forms? Did they learn 
the new script easily? Did the Emperor himself learn it? 
Was the merchant Lii Pu-wei competent in reading and 
writing? To all such questions there seems to be no answer 
for the historian today. All we know is that most teachers 
of the time were accustomed to using copies of one scripture 
or another in their possession. How complete those copies 
were we do not know, but by the end of the Chou dynasty 
no Ju could achieve repute as a teacher unless he possessed 
some tablets — tablets on which were recordings of traditional 
lore, ritual notes, divining oracles and reflections thereon, 
statements as to regulation behaviour for government serv- 
ants, etc. The more lively-minded among the teachers would 
write down their own reactions to the teachings they knew 
by heart, whether in the central Ju tradition or the Taoist 
or Legalist or what. The Ju of Lu State and Ch^i State were 
famous in late Chou times, and the authors of the Tso Chuan^ 
Kuo Yilj and Chan Kuo Ts^e must have used the services of 
quite a number of scribes. Away south in the Yangtze area 
Ch'ii Yiian could pour out his burdened heart in 374 im- 
passioned lines. At the same time it was easy to find "vulgar 

[ 249 ] 



GENERAL APPRAISALS 

Juj'^ pretentious men in big hats and voluminous gowns, with 
little understanding of what they had been taught. 

The Warring States period is forever famous for the gal- 
axy of highly individual thinkers whose (more or less) or- 
derly thinking took literary shape and was preserved by 
disciples. There were also those academies where these phi- 
losophizers could meet and argue. There was also Hsun 
Ch'ing, who seems to have first discovered that vital literary 
form, the discussion of a particular subject under its particular 
title. Hsiin Chang's theory of human nature drove him to 
see the necessity for an artificial disposition to be cultivated 
by continued practice in courtesy and unselfish renunciation, 
,, and his essays are found quoted at length in the "Record of 

:j Rites." Plainly, he was the father of the movement which 

j[ brought the proletarianization of Li in mid-Han times. In 

:i view of these well-attested phenomena no one can doubt that 

11 the lamp of literacy was well alight. In terms of thought 

cj and communication, the age that could produce the Mencius 

book and the more clearly original parts of Chuang Tzu was 
gifted with a mind which was well started on its climb to 
mature powers of ratiocination. Yet it is a far cry from that 
age to the intellectually more sophisticated age of Pan Ku 
and Chang Heng. It is not merely a matter of men of humble 
origin rising to be high-rank bureaucrats and being ennobled, 
but also of recognized ethical standards backed by a subtle 
understanding of the individual-in-society.^° 

Again, it is quite a far cry from Hsiin Chang's time back 

^^ To illustrate this development, T'i-wu Lun's career (see Hon Han S/iu, 
Biog"- 51) shows how a talented man of peasant origin could in Kuang 
Wu's reign be advanced from local office to high station. Nothing is recorded 
of his education, but his home was in the Ch*ang-an regional government 
area, so that a handy village school may be credibly assumed. From the 
other angle of social approach, Chang Heng had a rather younger con- 
temporary, Chang Kang (see Hou Han Shu^ Biog. 46), who came of a 
long-established noble family, and whose father, a man of outspoken char- 
acter, held high office. Chang Kang wore plebeian clothes and modeled him- 
self on Confucius in taking part in village confabulations. On one occasion 
he won over to law and order the rebels of a district in revolt against their 
avaricious official. Such a man would foster village education. 

[ 250 ] 



If 



GENERAL APPRAISALS 

to the days of young Confucius as he embarked on his career. 
There is no record of who his teacher was or what he, be- 
longing to a Ju family, had learnt from him, or of what 
sets of tablets he inherited, no record of other contemporary 
teachers and their pupils. It is triple-guarded tradition that, 
although he was never a state archivist, he had access to 
Lu State records and composed the "Spring and Autumn 
Annals." But with all his genius as a teacher and his pas- 
sionate concern for his followers he never took stylus and 
wrote down what he would have them to remember. The 
only possible conclusion seems to be that the idea of an 
individual on his own initiative expressing himself in cool 
and collected prose fashion had not occurred to anyone at 
that time. There is much to indicate that writing was still 
invested with a numinous aura, so that not till Confucius 
was dead did his followers feel they could and must write 
his sayings down. The rider to that conclusion is that formal 
education in that age could only have reached a rudimentary 
stage. 

To come back to Later Han, the spread of education meant 
for our two witnesses the spread of wen, i.e. civilized modes 
of living. They themselves are incontrovertible evidence of 
these more advanced cultural standards, mental, moral, and 
artistic. Their poems reveal a refinement of sentiment and 
principle which is unmistakeable. They also show the in- 
spiring effects on — at any rate some — intellectuals of "lifting 
up their eyes to the ends of the earth." And theirs was the 
age in which the Li Chiao, the code of manners and morals, 
intensified its hold on society, developing into a philosophy 
of Li in which old and new observances were compounded 
into one massive structure. To the Ju it was the very em- 
bodiment of wen. Yet to the modern mind, the more it is 
examined in detail, the more it appears as a restrictive, per- 
verting influence, fostering a pharisaical, hypocritical spirit. 
Whilst admitting some truth in this — a truth applying to 
all social codes of behaviour, ancient and modern — the Pan- 

[ 251 ] 



GENERAL APPRAISALS 

Chang evidence drives us to explore the opposite aspect, the 
enhancement of the individual and the means by which there 
is effected the alignment of practice with principle. The ex- 
ploration of this subject is beyond the scope of this final 
chapter, but two illustrations may be given to show the im- 
mense possibilities in this line of enquiry. 

The first is taken from Book X (init.) of the Analects. 
We find there illustrations of behaviour at court, and at first 
sight it is a fantastic picture of the Li Chiao at work. It is 
urged here that actually the author of those anecdotes was 
depicting absorbed self-expression in the conduct of duty. 
The second illustration is from the "Great Learning" {Ta 
Hsueh) in the "Record of Rites." At the outset there is a 
prime emphasis on the "cultivation of the self as the root 
process" j and this is linked with the "development of knowl- 
edge" and the "investigation of things," i.e. natural things. 
Any reader going through the "Great Learning" will dis- 
cover that this text-book of Li Chiao philosophy is concerned 
from first to last with the flowering of a noble personality. 

It is, therefore, assumed with some confidence, that the 
Li Chiao of Later Han set up standards of personal honor 
in all ranks of educated society — standards which survived 
the severest tests of later anarchy and disruption. On the 
other hand, there was one class of person in Later Han on 
which the Li Chiao failed to exercise its ennobling influence: 
the class of the emperors. A young heir-apparent — always, 
it must be remembered, on approval — was subjected to a 
severe scholastic discipline into the results of which the reign- 
ing Emperor would enquire. Yet, having been nurtured in 
his early years in the imperial harem, he was imbued with 
its enervating influence. As Emperor he was shut away from 
contact with the outside world, as Chang Heng's muffled 
dummy in the imperial carriage brings home to usj and all 
the time he was surrounded by flattering career-mongers. 
There was the endless routine of ritual duties in which he 
figured as a public god. Psychologically speaking, a man of 

[ 252 ] 



I 



GENERAL APPRAISALS 

outstanding character could rule circumstances to his will, a 
weakling could not. Hence, that dereliction of imperial au- 
thority which characterized the last fifty years of Later Han. 
From A.D. 89 on, the nine successive emperors all ascended 
the throne when they were boys or even babes in arms. 

8. Thd Sage-Kings 

The enquiry and appraisal here, it is hardly necessary to 
say, is limited to Han, to what ideas the Han intelligentsia 
before a.d. 120 had about the sage-kings. Since it was in that 
era that the Chinese people moved so vigorously out of their 
ancient northern environment and spread their dominion so 
widely, and in that era that their intelligentsia set their minds 
so methodically to work on the nature of legitimate sover- 
eignty, they could not but face up to their sacred past and 
rationalize it to the best of their ability. To what extent old 
legend became new myth or alternatively old myth became 
new legend, in relation to this or that sage-king, is probably 
one of the test questions in this field, but even that temptation 
is resisted in the discussion here. 

There are two points to be made about the references in 
these fu to the sage-kings. One is that there are no references 
to them at all in the two Ch^ang-an fu, but only in the Loyang 
fuy notably the references in V, 2 and VI, 12 which, if taken 
at their face value, point to both Pan Ku and Chang Heng 
believing in the whole series of sage-kings back to Fu Hsi. 
The other is the admirably logical emphasis Pan Ku laid 
(V, 2) on Kuang Wu as out-doing all the sage-kings put 
together, an emphasis paralleled in his own way by Chang 
Heng (VI, 12). All three features are surprising according 
to particular angles of consideration. 

First, how was it that, whether by intention or by inad- 
vertence, the creators of the two protagonists for Ch^ang-an 
omitted any reference in their speeches to the sage-kings? One 
weighty answer is, of course, that the Pan-Chang thesis about 
Ch^ang-an was that it embodied the true royal virtus much 

[ ^53 ] 



GENERAL APPRAISALS 

more in fosse than in esse. On the other hand, both authors 
knew perfectly well that in Emperor Wu's reign the much 
admired Tung Chung-shu had used his best reasoning pow- 
ers in demonstrating the march of history under the com- 
pelling power of cosmological forces, and that in his scheme 
the Han throne with its wen (high civilization) was in due 
succession to Hsia, Shang, and Chou, the regimes inaugu- 
rated by Sage-Kings Yii, Ch^eng T^ang, and King Wen." 
Also, Pan Ku at least was familiar with the fact that in the 
official eulogy in the "History of the Former Han Dynasty" 
Kao Tsu was represented as claiming descent from Yao. 
(Yet Pan Ku refers to Kuang Wu, not Kao Tsu, as linking 
his regime to Yao.) It looks very much as if in Pan Ku's 
day his divergence from Tung Chung-shu's theory in post- 
poning the revelation of the sage-king succession to Later 
Han, was accepted by the legitimist theoreticians. The point 
really is whether that theory could be and was adapted to 
present facts. There can be little doubt that it was. Thus, for 
example, in curiously illuminating fashion, Tung Chung-shu 
in his Ch'un ChHu Fan Lu (his only work which has sur- 
vived) shows no interest at all in any particular virtus of 
any sage-king. He is merely concerned with getting the suc- 
cession right, and in doing so he goes no further back in 
antiquity than the Hsia regime. True, he mentions Yao and 
Shun, but only in passing, and for the earlier sage-kings has 
only a tag reference, "Five Ti and Three Huang" ("August 
Ones"). None of these plays a part other than as links in 
his dynastic scheme. That was one type of rationalization, 
and a very prevalent type in mid-Han, incidentally throwing 

^^ One striking feature of Tung Chung-shu's theorizing was to include 
Confucius in the dynastic series as inaugurating a sort of half-in-Heaven, 
half-on-earth sage order of rule. One may suspect that he had in mind the 
bridging of the gap between Chou and Emperor Wu, but the connection 
is not clear. One thing is clear, that he was thinking a good deal in terms 
of the class "sage," whether ruler or not, and the influence of this generali- 
zation should be borne in mind. It comes out very prominently in the Po 
Hu T^ung's chapter "On Appellations" (see Tjan, of.cit., 232/6). 

[ 254 ] 



GENERAL APPRAISALS 

light on Wang Mang's use of the sage-kings in advancing 
his claim to establish his New Dynasty. 

It is submitted, therefore, that there were a large number 
of stories current about the sage-kings, as about other heroes 
of antiquity, and that not only Pan Ku and Chang Heng 
but many other scholars, too, sat rather lightly to them as 
a whole. Some legends were more credible than others which 
contained miraculous elements such as were told of the Yel- 
low Emperor, the part-Taoist, part Middle-Tradition hero. 
In either case many of the Ju were, like Tung Chung-shu, 
not seriously interested in them as individuals. They were 
eponyms, attached to this regime and the other, or, as the 
"Changes" Scripture affirmed, the semi-divine discoverers of 
society's civilized inventions. The paramount consideration 
was the procession of dynasties like the order of the seasons. 
If that were the case, then Chang Heng's reference to "the 
Five" and "the Two"^" is not to be taken as expressing his 
conviction that these far-off beings had actually existed. They 
were convenient pegs to which he could pin his argument 
and enforce his appeal. 

As for the romantic details of this and that sage-king, 
e.g. Yli who by his labors "wore the hair off his calves and 
shins" and passing the door of his house heard the cry of 
his new-born child but stayed not to see him — who knows 
whether Yii was originally a god or a man, whether myth 
was piled on legend or legend on myth? Whatever the sub- 
stratum of factual event, if any, it may have passed through 
a dozen transformations before it reached the final form we 
know, with its lesson that government is for the sake of the 
common man. Can the specialists trace any such intervening 
stages? But for this appraisal the question is a more clearly 
definable one. Take Chang Heng's Ssu Hsilan Fu with its 
endless recital of gods and goddesses whom he contacted 
round the world: did the poetic force of those sage-king 

^^ I have no theory to propose as to why Changf Heng substitutes "the 
Two" for "the Three," the formula common after Tung Chung-shu's day. 

[ ^ss ] 



GENERAL APPRAISALS 

images so appeal to him that their vigorous aptness made 
for him a sufficient substitute for actual historicity? The ques- 
tion may be unanswerable, but there is this to be said: since, 
whatever philosophers may argue, the images created by 
words so much are symbols of what they are supposed to 
represent, what was there to inhibit so scientifically minded 
a man as Chang Heng from adopting a highly poetic symbol 
as near enough to the actual truth? The dice were heavily 
weighted in the direction of such a belief. No one can read 
the "Changes" Scripture and then go on to Yang Hsiung's 
extension of its philosophy of symbols without realizing how 
profoundly an intelligent Ju was imbued with this quite 
arguable theory. After all, the Taoist Scriptures which Chang 
Heng knew maintained that the ultimate fact about the uni- 
verse was a mystery, from man's angle of approach, a noth- 
ingness from which sprang somethingness as a relative image. 
Along that line of speculation, what was the truth about the 
sage-kings: a faltering balancing of evidence, or the living 
image which struck an echo from your mind? 

9. Concerning Han ^^ Confucianism^^ and 
the Scriptures 

The critique discovered grounds for questioning the ap- 
propriateness of the Western Sinologists' term "Confucian- 
ism" for the "Great Tao," the "Middle Way," which grad- 
ually took shape in Han times. Then earlier in this chapter 
something more approaching the real teaching of Confucius 
of Lu was discovered as an unprecedented feature of the new 
Later Han political set-up. For the final appraisal here it 
must be confessed that no delineatory answers can be given 
to the questions which clamor for answers. For the historian 
the situation is one sown with paradoxes, as, for example, 
the fantastic conglomeration of ideas found in the report 
on the White Tiger Hall discussions and the fact that its 
official reporter. Pan Ku, burst out in his Loyang fu in ful- 
mination against the Scripture proof-text mongers who dom- 

[ ^5(> ] 



GENERAL APPRAISALS 

inated that conference. That the court's decision, some eighty 
years after Pan Ku's death, to inscribe the Scripture on stone 
pillars and its subsequent encouragement of Cheng Hsuan's 
commentaries point to a more intelligible situation may be 
true, but is beyond the scope of this enquiry. The only sug- 
gestion I venture to make is along the lines of generally 
accepted opinion today, namely that these developments point 
not only to the growing strength of the Ancient-Script party 
among the Ju but also to the central importance, in the eyes 
of the Scholars and the government, of getting the Scripture 
texts authoritatively fixed and their meanings authoritatively 
defined. There was no suggestion of this kind in the Pan- 
Chang age — and yet the plenary authority of the Scriptures 
as a whole was the main plank in the platform of Ju belief s.^^ 
In view of our relative ignorance of what the said Scriptures 
did or did not contain at that particular time, it is submitted 
that there is a pronounced x quality to the term "Confucian- 
ism." If we turn to Confucius of Lu as our main guide to 
its meaning, we may be able to advance up to a certain point, 
but then we are faced with the question as to which "Con- 
fucius" is the really historical one: the Confucius of the 
Analects, the Confucius of the Tso Chuan, of the "Changes" 
Scripture, of the biography in the "Record of History" (^Shth 
Chi), of the Modern-Script school with its emphasis on the 
"Kung Yang Amplification," of the "Filial Piety Scripture," 
or of the "Record of Rites," where he appears in so many 
passages discoursing at far greater length than he does in 
the Analects} Here again we are faced with a formidable x 
quality in the evidence at our command. 

^^ This statement, true in general, must not be taken as referring only 
to the "Five Scriptures." The Fo Hu T^ung contains a quantity of citations 
from other books of near-canonical status (cf. Tjan, of.cit., pp. 180-194). 
Among these the "Woof" (Wei) Books must be taken into account. They 
also figure as lending authority to the opinions recorded. Clearly some of 
these "Woof Books" were held in such esteem that their advocates, the 
Modern-Script party, were able to get near-Scripture recognition of them. 
The natural inference from that fact is that the "Warp" {Ching) Books 
did not constitute a closed canon in a.d. 93. 

[ 257 ] 



GENERAL APPRAISALS 

The historian today has one source of encouragement. It 
is that the patient work of critical scholarship has confirmed 
the Analects as the most nearly reliable account we have. It 
is second-hand evidence, third-hand, or even fourth-hand evi- 
dence, but its imperfectly correlated format and general co- 
herence of portraiture puts its value in a class by itself. 
Arguing from this angle of approach, one mid-Han and Later 
Han belief becomes highly questionable. It was that Con- 
fucius by his sage wisdom came to see that the records of 
sage wisdom of antiquity must be preserved for a more 
enlightened age when the scholars could prize them at their 
full value. Hence in general terms he was the transmitter 
of the Scriptures. With the strong backing of the Analects 
we can accept the "Odes" as a collection owing its existence 
to Confucius' enthusiasm. That, however, is a different prop- 
osition from believing that he regarded his collection as a 
Scripture in the Han sense. That he took a number of Chou 
court historical records — the Modern-Script and Ancient- 
Script devotees violently disagreed as to which and how 
many — and put them into final shape for the future as a 
"History" Scripture is hardly credible. That he in his last 
days was devoted to the "Changes" (as it was then) and 
added the amplifications of the "Wings," this has become 
more and more open to doubt. As for the "Rituals," specified 
in Later Han times as the / Liy Chou Li^ and constituent 
parts of the "Record of Rites," the grounds of the belief 
are dubious at every point. Not that Confucius was not a 
master of the Ju lore of ritual in his day, or that he did 
not possess tablets of notes on ritual practice. However, that 
he passed on an / Li Scripture as a whole, much less a Chou 
Li or any recognizable parts of the "Record of Rites" is 
definitely not credible. As for the "Spring and Autumn An- 
nals," although the Analects stresses his insistence on the rec- 
tification of terms, it knows nothing of such a work. Yet in 
Mencius (III B, 9) Confucius is stated to have been the 
inspired author and as saying "Those who know of me, will 

[ 258 ] 



GENERAL APPRAISALS 

it not be through the 'Spring and Autumn Annals'!" The 
problem is completely baffling, particularly because as an 
existent book it vanishes from sight until in Emperor Wu's 
reign it comes to light as a Scripture. It was acclaimed as 
such at the instigation of Tung Chung-shu, and along with 
it, wil/i very little delay ^ the Kung Yang Chuan. The as- 
tounding thing is that Scripture status was claimed for this 
amplification, and for its rival the Ku Liang Chuan^ and then 
later for the Tso Chuan^ a very different kind of book. For 
this last it was claimed that its author was a direct disciple 
of Confucius in his old age and assistant to him in composing 
the "Spring and Autumn Annals" (a very dubious proposi- 
tion).^* This direct link with "the Master" was not claimed 
for the two previously contested amplifications. 

It stares us in the face, this passion of the mid-Han Ju 
for a corpus of infallible Scriptures. It is as if amid all the 
Scholars' rivalries and controversies that was the fixed point 
in their minds: that insofar as Confucius came into their 
arguments he was a convenient pawn to be moved here and 
there in support of a main contention. Also, taking the Five 
Scriptures together, we realize that the scope of their com- 
bined ideologies goes far beyond what any Ju of the Spring 
and Autumn era could possibly have envisaged, even though 
he were one of Confucius' pre-eminent calibre. Also, through 
Pan Ku's and Chang Heng's eyes we have seen how pro- 
foundly the Ju had been impressed by the achievements of 
the Han monarchy, and how firmly they backed the new 
order of unity and headquarters control. That monarchy and 
order had to be buttressed, i.e. authenticated by metaphysical 
proofs. Those proofs were to be found in the tablets by means 
of which the teachers of the administrators and the learned 
had inducted them into the illuminating mysteries of learn- 
ing. Thus the Scriptures guaranteed the authenticity of the 
throne, and the throne guaranteed the authenticity of the 

^* See B. Karlgren, On the A iitJienttcity and Nature of the Tso CJiuan^ 
Goteborg:, 1926. 

[ 259 ] 



GENERAL APPRAISALS 

Scriptures. By the same logic applied to antiquity, the Scrip- 
tures authenticated the sage-kings, and the sage-kings authen- 
ticated the Scriptures.^^ To our modern intelligence this may 
be outrageous logic: to the Han dogmatist, a theologian in 
grain, it was doubly fortified reason. 

The question arises whether this Scripture-centric belief 
was a religion. To that this appraisal has but two suggestions 
to offer. One is the emphasis that Han religion and politics 
were indissolubly interrelated. It is urged not only as a peculiar 
feature of the Scholars' religion-cum-politics but also as a 
universal characteristic of all cultures, ancient and modern. 
The other suggestion is that although the Scholars' religion 
had such strikingly mundane elements, there was also a tran- 
scendent, mystical streak. This comes to view in Chang 
Heng's Loyang fu, where Scholar Anonymous expresses him- 
self in markedly Taoistic terms. 

The first impression in Western Sinology was that "Con- 
fucianism" and "Taoism" were basically antagonistic views 
of life and reason and that the glaringly superstitious Tao 
experts in Former Han times degraded Taoist beliefs and 
discredited them in the eyes of the scholars, and so Con- 
fucius' more rational views became the Great Tradition of 
the Chinese people. More recent scholarship has discredited 
this over-simple view. Not that Emperor Wu's shamans did 
not disgust the more intelligent scholars, nor that the Chuang 
Tzu^s socially nihilistic philosophy of the individual did not 
strike them as unpractical. But there was a magic in the pure 
transcendentalism of the Tao Te Ching and the Chuang Tzu 
which the more sensitive Ju, those less obsessed with ambi- 

^^ This kind of reasoning is seen very clearly in the Lun Hengy e.g. ch. 
28, c. I (A, Forke, T/ie Lun Heitg, Philosofhical Essays of Wang Ch^tmgy 
vol. I, Kelly and Walsh, 1907, p. 4.52). There we see Wang Ch'ung him- 
self under this influence, although at the same time he maintained that the 
Scriptures had suffered irreparable damage, i.e. had been recovered in 
Han times in very imperfect condition. In ch. 28, c. 2 (Forke, of.cii.y 
vol. II, p. 238) he unintentionally reveals the dominant belief: "Supposing 
that the Five Scriptures had come undamaged from Confucius' disciples j 
then they might be said to be a perfect whole and could be trusted." 

[ 260 ] 



GENERAL APPRAISALS 

tion, could not resist. One key to the mid-Han attitude is 
to be found in Liu Hsin's Catalogue, where the Taoist sec- 
tion is placed next to the "Confucianist" section of the philos- 
ophers' works, and that and the other succeeding sections are 
all appraised as contributory to the elucidation of truth. Here 
again is a sign of the belief in the power of the written word, 
a semi-sacred word, throwing light, if only by contrast, on 
the completely sacred word of the Scriptures. So when we 
turn to the central theme of the four fu, the Han imperial 
virtus and its mystic life-engendering powers, it is plain that 
Later Han Ju beliefs owed much to Taoism. 

10. Suggestions for Systematic Research 

The critique has been at considerable pains to estimate 
the intelligence quota of Pan Ku and Chang Heng, and to 
analyse out their respective thought processes. At the be- 
ginning of this chapter an attempt was made to estimate the 
results. Since both authors succeeded in being remarkably 
self-revelatory, there is ground for the hope that the results, 
patchy as they are, may be of use to Han historians and even 
inter-cultural historians. On the other hand, the actual sphere 
of observation has been so restricted that obviously further 
studies on a wider basis are needed. In the case of Pan Ku 
there is the field of enquiry afforded by his eight other works 
in the Wen Hsiian^ none of them of importance comparable 
with his "F^ on the Two Capitals," but nonetheless needing 
more careful scrutiny than I have had time and energy to 
give. Also, there are various old and new studies of Pan 
Ku's worth as a historian, particularly by Chinese critical 
scholars. From these, owing to my recent more narrowed 
conditions of study, I have been unable to profit. 

With regard to Chang Heng, in the Wen Hsuan there 
are four other items, including his ^'Fu on the Southern 
Capital." A preliminary study of this left me with surprise 
at its lack of significance, and it dropped out of view in my 
later studies. His Ssu Hsilan Fu is of supreme importance 

[ 261 ] 



GENERAL APPRAISALS 

for the understanding of him and would vastly repay the 
most painstaking study, both in relation to the sage-kings and 
to his world outlook. There are also various items in Yen 
K^o-chiin's compendium Ch^uan Han Wen ("Complete Lit- 
erature of the Han") which are necessary for the clearer 
understanding of Chang Heng's attainments as a scientist. 
Both Pan Ku and Chang Heng need to be rediscovered as 
products of the conventional Scripture-based education com- 
bined with the inspiration of the mid-Han literary up-rush. 

Then for the sociological and ideological anthropologists 
surely there are valuable data in that great virtus theory 
which Pan Ku and Chang Heng, the historian and the scien- 
tist, make the central theme of their poetic discourses: the 
concept of te^ reaching so far back into the dim yet turbulent 
life of early animistic tribal society. Here it appears in all 
its glory of civilized, ethical application, the key to Later 
Han's consciousness of a world mission of peace and friendly 
cooperation under the guidance of Heaven. Does it not in 
its lights and shades stand in ancient history on a par with 
Constantine's espousal of the Christian religion and Asoka's 
espousal of the Buddhist cause? 

As to that literary up-rush, it above all requires the most 
painstaking investigations. It is off the main line of historians' 
researches today and seems to be regarded as a very special- 
ized arcanum into which only rather antiquarian-minded lit- 
erary critics of the purest water need to probe. Yet one plain 
fact about those momentous two centuries of Former Han 
was the acceptance of the First Ch^in Emperor's experimental 
new script. The experiment was a success, both for govern- 
mental purposes and as a tool in the hands of the nation's 
educators. Apparently all parties took to it and proceeded 
to make themselves highly proficient in it. The pre-Han lit- 
erature, including the Scriptures, such as they were, were re- 
inscribed in it, so that by the time Liu Hsin's Catalogue 
was made, scrolls occur in it for new as well as old manuscripts. 

The significance of this revolution can only be realized 

[ 262 ] 



GENERAL APPRAISALS 

in the light of modern studies in the ancient script and 
phonetics. With much that still remains dark and uncertain 
there are two features of the end-of-Chou age which may 
be taken as sufficiently clear and certain. One is that in the 
Chou centuries the substitution of the phonetic principle for 
the original pictographic and ideographic principle had slowly 
but surely come to pass. The other is that this process was 
not autocratically regulated, and when the rival states (in- 
cluding Ch^u in the then far south) came to their full power, 
strong divisive cultural influences were at work. That script, 
the hsiao chilan ("small seal" script, or whatever it may 
have been), was subjected to regional variations. To that 
was added the confusion of specialized new meanings to old 
script forms, a process enhanced by the lively schools of radi- 
cal philosophizing. The written language was altogether in 
bad shape, in addition to which the stylus and the wooden 
tablet were clumsy laborious media of communication. 

The Ch^in First Emperor by brute force substituted his 
rule for the effete Chou hegemony, and he and his prime 
minister Li Ssu realized that without a universally recog- 
nised script universal government could not but be hampered 
at every turn. They promptly made a new and simpler script. 
What mental processes activated Li Ssu and his assistants 
must in the absence of evidence be largely a matter of con- 
jecture, stimulated by the patient collection of indirect clues. 
But the main fact is clear that their new script, being aimed 
at producing clarity of communication and speed in writing, 
did achieve this end, and that it did so by enlarging the scope 
of the Chou phoneticizing trends. The greater use of phoneti- 
cally based terms was coupled with the use of phonetic loans 
(^chia chieh). 

It is impossible now to trace what struggles teachers and 
students had over this new departure in the written word. 
There is, however, one significant item in Liu Hsin's Cata- 
logue. It is the assumption that from the beginning of writ- 
ten works of public interest, an imprimatur had to be issued 

[ 263 ] 



GENERAL APPRAISALS 

by some department of state. That would seem to indicate 
that in Former Han some sort o£ headquarters pressure was 
exercised in obtaining conformity in the use of terms, i.e. of 
scribal conformity. In any case, by the end of the Former 
Han Dynasty the ancient awe over the written word with 
its mana-imbued symbols had changed into intelligent ap- 
preciation of wen (patterned language) as a noble art of the 
mind (/isin). A sharp distinction was drawn between fi (rough 
notes) and the studied expressions of considered thought. 
From this angle of approach Tung Chung-shu's seventeen 
chuan of philosophizing in his Ch^un ChHu Fan Lu becomes 
an exciting document, a dated example of literary capabilities 
on the way. Ssu-ma Ch^en's "Record of History," a docu- 
ment of slightly later date, running to 130 chilan^ is even 
more exciting from the literary angle, for by his mastery in 
sentence progression and chapter division literature on the 
grand scale is on the way, and the door opened for Pan 
Ku's masterly advances in the writing of history. Even more 
exciting, as a contemporary event to Ssu-ma Ch^en's achieve- 
ment, is Ssu-ma Hsiang-ju's amazing revivification of the 
old ju genres. He transformed the patterned forms of Ch^u 
Yuan's impassioned introspection, of the Ch^u Tz^u^s animis- 
tically-inspired reflections, and of Hsiin Chang's and other 
Scholars' didactic poetry, converting them into a medium 
for delineating large-scale objects of observation, and he did 
this in an overall vein of irony strengthened by the use of 
dramatis fersonae. There is no need to cite Pan Ku's other 
instances of the new literary capacities exhibited in the later 
Ch^ang-an court. It is enough to cite Pan Ku and Chang 
Heng, the architects of those four grandly conceived poetic 
structures revealing the discriminate features of the two Han 
capitals and regimes, to realize what heights of communi- 
cative power that recreated script made possible. 

The whole field of dateable mid-Han literature requires 
reexamination from the angle of this central idea — the prose 
as well as the poetry, the state papers and reports of con- 

[ 264 ] 



GENERAL APPRAISALS 

ferences, the memorials and warnings (probings), the en- 
comiums and bewailing condolences, not omitting the styled 
products of courtly badinage and riddle-making. Anywhere 
may be found grist to the explorers' mill. In particular, since 
it was in the period that the Yin-Yang cosmology really 
asserted its power over Scholar minds, stirring them to all 
sorts of ratiocination, close attention needs to be paid to the 
emergent forms of speculation. For example, from it stemmed 
a rationale of history, of what ancient history must logically 
have been: extremely convincing to minds under the sway 
of ineluctable processes, birth, growth, decline, and death. 
Also its effect on language awaits more rigorous investigation: 
the growth of dioptic expressions, "looking up . . . looking 
down," "the inner . . . the outer" (point of view), "referring 
to the past . . . referring to the present." They reveal an 
analytical consciousness in regard to angles of consideration 
on a subject of reflection. They are the more impressive be- 
cause the same ability is disclosed in relation to the aspects 
of Nature, heaven and earth, summer and winter, the curved 
and the rectilinear, the dark and the light, the fine and the 
rough. Also the same complementary distinctions were worked 
out in relation to man, man in the family, man in society 
and the nation.^^ Whether the distinctive qualities of the 
new script aided or retarded the development of these new 
powers of the mind is awaiting research, and along with that 
must go research into the effects of the Yin-Yang mind on 
the structure of sentences and the tightening of discourse. 
Finally, evidence of this power to drive the mind in double 
harness may seem to be forthcoming from late-Chou litera- 
ture, evidence at any rate of rudimentary forms of it. There 
is, however, the inescapable doubt whether we can be sure 

^^ For further ideas on this subject, see my paper, "The Epistemological 
Method in Chinese Philosophy," in Essays in East-West P/i'ilosofhy^ Uni- 
versity of Hawaii Press, 195 i. A good example of "double-harness" reason- 
ing is to be found in Liu Hsieh's Wen Hsin Tiao Lung, ch. 1, a transla- 
tion of which appears as Appendix 2 in my Art of Letters, BoUingen Series, 
New York, 1951. 

[ 265 ] 



GENERAL APPRAISALS 

of any such book having come through the Han era without 
undergoing some change, even if only a small one. To refer 
only to Liu Hsiang and his colleagues, they took their duties 
of redaction very seriously, and we know that they had dif- 
ficulties. There was also the transcribing of old-script docu- 
ments into the new script, and we know that many mid-Han 
Scholars had a passion for demarcation of the text, as also 
for composing amplifications, elucidatory of the meaning as 
they saw it. The doubts thus created cannot be entirely re- 
solved by citation of recorded instances of Erudites and others 
being severely punished for tampering with sacrosanct writ- 
ings. Modern critical scholarship, stirred by the Ch^ing Dy- 
nasty learned questionings, has continued on the long road 
of linguistic and stylistic enquiry. The suggestion here is that 
from more detailed, more exact knowledge of the new styles 
of language selection and sentence structure which spring to 
life in mid-Han times there might emerge new powers of 
discrimination in relation to the Scriptures. In particular it 
is urged that the most likely Scripture to produce results 
from the application of this technique is the "Record of 
Rites," known as a compilation of mid-Han Ju. That the 
intricate weaving of old and new material in many of its 
books and passages would, of course, create problems, in some 
cases insoluble ones, would naturally be expected. But the 
new style, the "contemporary style," very much laid its hands 
on time-honored phrases and disciplined them into line with 
its passion for short clear-cut clauses and sentences in which 
one statement is made to impinge sharply on its fellow state- 
ment. By this means the twisted cue of Scripture studies might 
be untwisted somewhat, the cloth untrue be levelled in some 
texts, and the wobbling ellipsis of dating documents become 
more amenable to direction. 



[ 266 ] 



COLLEGE LIBRARY 

Date Due 
Due Returned Due Returned 


' 1 * - 








Mf 

















































































































































































































UNIVERSITY OF FLORIDA 



3 1262 05285 1408 



\^PS> 






\